Note: Descriptions are shown in the official language in which they were submitted.
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
SYSTEM AND METHOD FOR TEXT CATEGORIZATION AND
SENTIMENT ANALYSIS
CROSS-REFERENCE TO RELATED APPLICATIONS
Application Date Filed Tide
No.
Current Herewith SYSTEM AND METHOD FOR TEXT
application CATEGORIZATION AND SENTIMENT
ANALYSIS
Is a PCT filing of; and claims priority to:
16/794,162 Feb. 18, 2020 SYSTEM AND METHOD FOR TEXT
CATEGORIZATION AND SENTIMENT
ANALYSIS
which is a continuation of
16/283,447 Feb. 22, 2019 SYSTEM AND METHOD FOR TEXT
Patent: Issue Date: CATEGORIZATION AND SENTIMENT
10,565,244 Feb. 18, 2020 ANALYSIS
the entire specification of each of which is incorporated herein by reference.
BACKGROUND OF THE INVENTION
Field of the Art
[001] The disclosure relates to the field of information processing, and more
particularly to
the field of analyzing provided text representing conversations to analyze
them for sentiments
and performing categorizations that make a response to expressed sentiment
actionable.
Discussion of the State of the Art
[002] It is currently commonplace in textual analysis, to use regular
expressions with
dictionaries of words and databases of common or anticipated nouns, to perform
simple
lookups and pattern-matches to loosely categorize subject matter and sentiment
during a
conversation or from a text sample provided to a given system. This may be
done to analyze
the sentiments of people communicating on message boards on the Internet, or
to gauge
them during text conversations with chatbots online such as for customer
service purposes, or
this may be done for information collecting purposes for law enforcement and
human
resources organizations, and even to detect unwanted messages in services such
as email and
1
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
text messaging services as well as the sentiment expressed in conversations
with contact center
agents on various topics, e.g. relating to sales and service.
[003] While current efforts for computing categorization and sentiment from
text may be
able to gauge user sentiment with some degree of accuracy some of the time,
there is
considerable lack of detail and a considerable margin for error in many cases
using current
simplistic systems. Emails may be sometimes erroneously gauged as spam, texts
or messages
on social networks and message-boards may be erroneously flagged for
moderation or
deletion, or their content may be inaccurately gauged for users searching for
specific forms of
content.
[004] What is needed is a system which will analyze the sentiment of a piece
of
conversational text with high accuracy (precision and recall) and do so within
the context of
user-defined categories and to monitor the change over time of the
distribution of textual data
that falls within each category together with its sentiment. Furthermore, a
system is needed
that can also discover the emergence of new categories automatically without
them having to
be pre-defined.
SUMMARY OF THE INVENTION
[005] Accordingly, the inventor has conceived, and reduced to practice, a
system and
method for improved categorization and sentiment analysis.
[006] A system for categorization and sentiment analysis is disclosed,
comprising: a chunk
parser comprising at least a plurality of programming instructions stored in a
memory and
operating on at least one processor of a computer, wherein the programmable
instructions,
when operating on the at least one processor, cause the at least one processor
to: receive input
in text form; break the text into chunks of text comprising words and phrases;
and compute
sentiment on the text at the chunk level; and a deterministic rules engine
comprising at least a
plurality of programming instructions stored in a memory and operating on at
least one
processor of a computer, wherein the programmable instructions, when operating
on the
processor, cause the processor to: categorize the text into pre-defined
categories using regular
expression rules and store the categorization; if no regular expression rule
is matched,
forward the chunked text to a semantic similarity engine; and a semantic
similarity engine
comprising at least a plurality of programming instructions stored in a memory
and operating
on at least one processor of a computer, wherein the programmable
instructions, when
operating on the at least one processor, cause the at least one processor to:
receive chunked
2
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
text; represent each chunk of text as a vector embedded in a high dimensional
space
representing semantic characteristics of the chunked text; categorize the
chunked text into
pre-defined categories using a threshold semantic similarity distance
(hypersphere radius)
from any of a set of pre-defined anchor word sequences for each category; and
if no
sufficiently close match is found to any pre-defined category anchor word
sequences, forward
the chunked text with embedded vector dimensions to a semantic cluster
discovery engine;
and a semantic cluster discovery engine comprising at least a plurality of
programming
instructions stored in a memory and operating on at least one processor of a
computer,
wherein the programmable instructions, when operating on the at least one
processor, cause
the at least one processor to: receive chunked text with embedded vector
dimensions;
determine additional new categorizations for the chunked text by analyzing the
text for
contextual associations using a semantic clustering analysis and store the
additional cluster
categories; and a category and sentiment analysis engine comprising at least a
plurality of
programming instructions stored in a memory and operating on at least one
processor of a
.. computer, wherein the programmable instructions, when operating on the at
least one
processor, cause the at least one processor to: receive the input text;
retrieve the
categorizations of the chunked text; analyze the sentiment of categories of
interest to a user of
the system; and output the results of the analysis to the user in the form of
text, graphics, or
both.
.. [007] A method for categorization and sentiment analysis is disclosed,
comprising the steps
of: receiving input in text form; breaking the text into chunks of text
comprising words and
phrases, using a chunk parser; computing sentiment on the text at the chunk
level, using a
chunk parser; categorizing text into pre-defined categories using regular
expression rules and
storing the categorization, using a deterministic rules engine; forwarding the
chunked text to
a semantic similarity engine if no regular expression rule is matched, using a
deterministic
rules engine; representing each chunk of text as a vector embedded in a high
dimensional
space representing semantic characteristics of the chunked text, using a
semantic similarity
engine; categorizing the chunked text into pre-defined categories using a
threshold semantic
similarity distance from any of a set of pre-defined anchor word sequences for
each category,
.. using a semantic similarity engine; forwarding the chunked text with
embedded vector
dimensions to a semantic cluster discovery engine if no sufficiently close
match is found to a
pre-defined category anchor word sequence, using a semantic similarity engine;
determining
additional new categorizations for the chunked text by analyzing the text for
contextual
3
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
associations, using a semantic cluster discovery engine; storing the
additional cluster
categories, using a semantic cluster discovery engine; retrieving the
categorizations of the
chunked text, using a category and sentiment analysis engine; analyzing the
sentiment of
categories of interest to a user of the system, using a category and sentiment
analysis engine;
and outputting the results of the analysis to the user in the form of text,
graphics, or both,
using a category and sentiment analysis engine.
BRIEF DESCRIPTION OF THE DRAWING FIGURES
[008] The accompanying drawings illustrate several aspects and, together with
the
description, serve to explain the principles of the invention according to the
aspects. It will be
appreciated by one skilled in the art that the particular arrangements
illustrated in the
drawings are merely exemplary and are not to be considered as limiting of the
scope of the
invention or the claims herein in any way.
[009] Fig. 1 (PRIOR ART) is a block diagram illustrating an exemplary
architecture of a
typical regular expression system and functionality to analyze categories from
conversations.
.. [010] Fig. 2 is a method diagram of a system to analyze sentiment and
categories in
conversations from users using new text analysis techniques, and specific
sequence of steps
taken to begin an initial parsing and analysis of text to ready it for further
analysis according
to a preferred embodiment.
[011] Fig. 3 is a diagram of differing techniques for assigning an input word
sequence to a
pre-defined category using a semantic distance "hypersphere" approach.
[012] Fig. 4 is a block diagram of a system architecture for semantic
sentiment analysis,
according to a preferred aspect.
[013] Fig. 5 is a block diagram of core components in a deterministic rules
engine,
according to a preferred aspect.
[014] Fig. 6 is a block diagram of core components in a semantic similarity
engine,
according to a preferred aspect.
[015] Fig. 7 is a block diagram of core components in a semantic cluster
discovery engine,
according to a preferred aspect.
[016] Fig. 8 is a block diagram of core components in a category and sentiment
analysis
engine, according to a preferred aspect.
4
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
[017] Fig. 9 is a block diagram illustrating an exemplary hardware
architecture of a
computing device.
[018] Fig. 10 is a block diagram illustrating an exemplary logical
architecture for a client
device.
[019] Fig. 11 is a block diagram showing an exemplary architectural
arrangement of
clients, servers, and external services.
[020] Fig. 12 is another block diagram illustrating an exemplary hardware
architecture of a
computing device.
[021] Fig. 13 is a method diagram illustrating key high-level functionality
and steps taken in
the operation of a category detection and sentiment analysis system, according
to a preferred
embodiment.
DETAILED DESCRIPTION
[022] The inventor has conceived, and reduced to practice, a system and method
for
improved categorization and sentiment analysis.
[023] A large volume of textual information exists online, particularly where
customers
express opinions about products and companies. Furthermore, advances in
automatic speech
recognition have provided another source of textual information by allowing
the conversion
of voice conversations into text in near real-time. In the case of call
centers, this means that
phone calls can be converted to text and analyzed for emotional content. To
the extent that
businesses can understand the emotional content of the customer's expressions,
they can take
action to optimize the business/customer relationship.
[024] The analysis of the emotional content of text-based information can be
broadly
separated into two forms: sentiment analysis and emotion analysis. Sentiment
analysis is the
simpler of the two. It attempts to identify the writer's attitude on a one-
dimensional scale
(e.g., positive, neutral, negative). Emotion analysis is more complex and
difficult, as it requires
a multi-dimensional analysis of the emotions the writer is expressing in the
text (e.g. anger,
sadness, fear, etc.). A non-exclusive list of other types of textual analysis
includes: semantic
analysis (the analysis of the writer's meaning and intent), topic extraction
(the analysis or
summarization of the overall topic of conversation in a writing), ideological
analysis (placing
the writer's opinion within a certain ideological classification), and
quantitative analysis (how
many times a certain thing is mentioned).
5
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
[025] Existing text analysis systems are inefficient and often not very
effective; to obtain an
acceptable level of accuracy, they generally require a large body of words and
phrases to be
created and manually tagged with emotional information by humans. This process
is slow,
cumbersome, and expensive and each body of words and phrases is limited to use
in the
.. specific context or field for which it was created. Even with a large body
of words and
phrases, many of the approaches currently used provide less than optimal
results even with a
large body of words and phrases.
[026] Another challenge is the lack of emotionally labelled text data in a
particular business
domain. A specific business domain may have a unique lexicon (vocabulary)
relating to
specific products and services and business-specific patterns of text. These
special words
resulting to the business would rarely be used with emotion in general English
usages outside
of the business.
[027] A taxonomy of desired categories of entities that are involved in a
specific business are
not hard for a subject matter expert to define in advance. For example,
Products, People,
Places and Processes. However, assigning entities or short phrases that occur
in the text to
each of these categories in terms of a set of fixed rules is tedious because
the names of all
people, products, places and processes would have to be specified in advance
together with
the rules to associate them to each specific target category. For example,
"Robert" to People,
"Mary" to People, "caller" to People, "agent to People" etc. Moreover, as
people and
products change over time the assignment rules need to be kept up to date.
[028] One approach to simplify this process may be by first specifying only
one common
example of a word or word sequence that is typical for each category, called
an "anchor"
word or word sequence. One may then leverage a pre-trained word sequence
embedding
model that has been trained on common general language usage to identify
incoming words
.. or word sequences which are semantically close to these predefined anchors.
In this way one
can then define an entire entity category by giving a single example of an
instance of it
instead of an exhaustive list with rules to map each word to the category also
taking into
account variations like plurals or even common mis-spellings. Some categories
may be
homogenous, consisting of entities that fall very close to each other in
semantic space. For
example, if a company only sells fruit as their products, it is possible to
define the whole
product category through the use of a single "anchor" entity expressed as
("banana", 0.8). It
is then possible to assign any incoming word (or word sequence) to the product
category if the
embedding vector of the incoming word (or sequence) is within a (semantic)
distance of 0.8 in
6
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
the embedding space from the pre-specified anchor word "banana". The use of a
real
number distance threshold like 0.8 allows to define the "tightness" around the
anchor word
phrase. A small value of the tightness results in only accepting words which
are very close to
"banana" in meaning, whereas a large value will match broader names of fruit,
and if the
"tightness" is set too large it will include other entities increasingly
dissimilar to fruit and
ultimately entities that are not fruit. In the case where a category consists
of very
heterogenous entities, e.g. a company whose products are fruit and musical
instruments, then
the system can be given "anchor" words for each subgroup within a category,
such as
[("banana",0.8), ("piano",0.8)]. It is also possible there may be a very
domain-specific use of
language around a company where product names differ broadly from their common
use in
English. For example, "apple" may refer to the name of a company rather than a
fruit. In
such a case, the system may apply exceptional rules for pattern matching of
words to
categories before the text is then passed to the approach using semantic
similarity based on
common usage.
[029] There are deep learning models that have been trained on large datasets
that have
been labelled by humans, such as "Recursive Deep Models for Semantic
Compositionality
Over a Sentiment Treebank" by Socher et al. However, the aim of these models
is to improve
the state of the art for sentiment detection on single sentences or phrases
compared to human
judges. The current invention concerns the task of how sentiment analysis is
combined with
categorization of sequential portions of text in such a way that: sentiment is
tracked about
pre-defined categories in the business (e.g. people, process, products,
places, etc.). If no match
is found to any pre-defined category, new emerging categories (non-predefined)
can be
automatically discovered and tracked dynamically, i.e. the system adapts
category discovery
over time. The pre-defined categories can be easily configured for different
businesses
(business "taxonomies") by non-experts (non-data scientists). The system
requires little or no
training data so that the system can be put into production from day 1 of
deployment in a
new environment. The system avoids the need for complex rule and pattern
matching
configuration for new environments; and the system has a highly flexible but
logical sequence
of categorization techniques following the sequence (i) fixed rule, followed
by (ii) soft semantic
hypersphere, followed by (iii) cluster discovery.
[030] The system uses semantic vector embedding spaces to find soft semantic
matching to a
pre-defined category by way of sentiment distance, but is implemented in a
modular way
where the specific vector space embedding module can be swapped from one type
to another
7
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
(e.g. sentence2vec, "universal sentence encoder", "paragraph2vec", "Paragraph
Vector" or
even simpler non sequence that use only the distributional similarities of
single words like
word2vec, or other semantic vector techniques to be invented in the future
[031] The system further allows the method of sentiment analysis on the
portion of text to
be swapped with another in a modular way.
[032] One or more different aspects may be described in the present
application. Further,
for one or more of the aspects described herein, numerous alternative
arrangements may be
described; it should be appreciated that these are presented for illustrative
purposes only and
are not limiting of the aspects contained herein or the claims presented
herein in any way.
One or more of the arrangements may be widely applicable to numerous aspects,
as may be
readily apparent from the disclosure. In general, arrangements are described
in sufficient
detail to enable those skilled in the art to practice one or more of the
aspects, and it should be
appreciated that other arrangements may be utilized and that structural,
logical, software,
electrical and other changes may be made without departing from the scope of
the particular
aspects. Particular features of one or more of the aspects described herein
may be described
with reference to one or more particular aspects or figures that form a part
of the present
disclosure, and in which are shown, by way of illustration, specific
arrangements of one or
more of the aspects. It should be appreciated, however, that such features are
not limited to
usage in the one or more particular aspects or figures with reference to which
they are
described. The present disclosure is neither a literal description of all
arrangements of one or
more of the aspects nor a listing of features of one or more of the aspects
that must be present
in all arrangements.
[033] Headings of sections provided in this patent application and the title
of this patent
application are for convenience only, and are not to be taken as limiting the
disclosure in any
way.
[034] Devices that are in communication with each other need not be in
continuous
communication with each other, unless expressly specified otherwise. In
addition, devices that
are in communication with each other may communicate directly or indirectly
through one
or more communication means or intermediaries, logical or physical.
[035] A description of an aspect with several components in communication with
each other
does not imply that all such components are required. To the contrary, a
variety of optional
components may be described to illustrate a wide variety of possible aspects
and in order to
8
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
more fully illustrate one or more aspects. Similarly, although process steps,
method steps,
algorithms or the like may be described in a sequential order, such processes,
methods and
algorithms may generally be configured to work in alternate orders, unless
specifically stated
to the contrary. In other words, any sequence or order of steps that may be
described in this
patent application does not, in and of itself, indicate a requirement that the
steps be
performed in that order. The steps of described processes may be performed in
any order
practical. Further, some steps may be performed simultaneously despite being
described or
implied as occurring non-simultaneously (e.g., because one step is described
after the other
step). Moreover, the illustration of a process by its depiction in a drawing
does not imply that
the illustrated process is exclusive of other variations and modifications
thereto, does not
imply that the illustrated process or any of its steps are necessary to one or
more of the
aspects, and does not imply that the illustrated process is preferred. Also,
steps are generally
described once per aspect, but this does not mean they must occur once, or
that they may
only occur once each time a process, method, or algorithm is carried out or
executed. Some
steps may be omitted in some aspects or some occurrences, or some steps may be
executed
more than once in a given aspect or occurrence.
[036] When a single device or article is described herein, it will be readily
apparent that
more than one device or article may be used in place of a single device or
article. Similarly,
where more than one device or article is described herein, it will be readily
apparent that a
single device or article may be used in place of the more than one device or
article.
[037] The functionality or the features of a device may be alternatively
embodied by one or
more other devices that are not explicitly described as having such
functionality or features.
Thus, other aspects need not include the device itself.
[038] Techniques and mechanisms described or referenced herein will sometimes
be
described in singular form for clarity. However, it should be appreciated that
particular
aspects may include multiple iterations of a technique or multiple
instantiations of a
mechanism unless noted otherwise. Process descriptions or blocks in figures
should be
understood as representing modules, segments, or portions of code which
include one or
more executable instructions for implementing specific logical functions or
steps in the
process. Alternate implementations are included within the scope of aspects in
which, for
example, functions may be executed out of order from that shown or discussed,
including
substantially concurrently or in reverse order, depending on the functionality
involved, as
would be understood by those having ordinary skill in the art.
9
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
Definitions
[039] "Artificial intelligence" or "AI" as used herein means a computer system
or
component that has been programmed in such a way that it mimics some aspect or
aspects of
cognitive functions that humans associate with human intelligence, such as
learning, problem
solving, and decision-making. Examples of current AT technologies include
understanding
human speech, competing successfully in strategic games such as chess and Go,
autonomous
operation of vehicles, complex simulations, and interpretation of complex data
such as images
and video.
[040] "Corpus" as used herein means a collection of natural language words,
phrases, or
both. The words and phrases in the corpus may optionally by tagged with
emotional
information for use in performing textual analysis.
[041] "Emotion analysis" and "Emotion classification" as used herein have the
same
meaning, and refer to the automatic detection of the emotions that a writer is
experiencing in
relation to an entity from an analysis of the text of the writing. Emotion
analysis is multi-
dimensional, and thus more complex and difficult, than one-dimensional
sentiment analysis.
[042] "Emotional content" as used herein means the emotions that a writer is
experiencing
in relation to an entity.
[043] "Lexicon" as used herein means vector information for each word or
phrase in the
corpus that provides information regarding meaning, associations,
relationships, sentiment,
emotion, and other information useful for performing textual analysis.
[044] "Machine learning" as used herein is an aspect of artificial
intelligence in which the
computer system or component can modify its behavior or understanding without
being
explicitly programmed to do so. Machine learning algorithms develop models of
behavior or
understanding based on information fed to them as training sets, and can
modify those
models based on new incoming information.
[045] "Sentiment" as used herein means the writer's attitude toward an entity.
The
sentiment (sometimes called "polarity" or "valence") of a text is typically
classified into a
three-value scale (positive, neutral, negative) or a five-value scale (very
positive, positive,
neutral, negative, very negative).
[046] "Sentiment analysis" as used herein means the automatic detection of a
writer's
attitude toward an entity from an analysis of the text of the writing. The
goal of sentiment
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
analysis is to determine the sentiment expressed by the writer of the text, or
phrases, and
sentences within the text.
[047] The term "categorization" is used here to mean the assignment of a
portion of text
concerning entities or collections of entities of interest to a business or
organizational category
that can be either very fine grained, broad, grouped or also hierarchical in
nature. The
category can cover any type of physical or conceptual entity (e.g. people,
products, process,
places, digital assets, etc.). This contrasts with "topic extraction" the
purpose of which is to
simply summarize a larger amount of text to simply understand what it is
about. Having fine
grained categories is important in order to make the observed sentiment
actionable. The
category defines what the sentiment is being expressed about so that category
can be
independently managed and improved.
Conceptual Architecture
[048] Fig. 1 (PRIOR ART) is a block diagram illustrating an exemplary
architecture of a
typical regular expression system and functionality to categorize
conversations. A possible
series of conversation textual data 110 is present, consisting of at least one
turn in at least one
conversation 111, and possibly other turns from other people in a conversation
112, and
possibly containing more than one conversation 113. In this prior art a "turn"
is one side of a
party in a conversation and can be quite long in length. Such data is read by
a conversation
analyzer using regular expressions 120 on the words used in such
conversational data 110.
Examples of regular expressions, without regard to a specific syntax or
methodology, may
include such things as searching a sentence for nouns that will e.g.
categorize a turn as
relating to a category "Products" by e.g. explicitly mentioning an exact name
of a product. In
addition, qualifiers such as "good," "bad," "great," "enjoyed," or other such
emotional
qualifiers, as is both commonplace in such sentiment analyzers and commonplace
in the
.. technical specifications of regular expressions, which are designed to find
basic patterns in
textual data rapidly. Such things that may be located out of a textual piece
of data 110 may
include product names or brands 130, names or titles of people 140, names of
processes 150,
and names or possible descriptions of locations and physical places 160. Such
pattern and
word matching is commonplace and easy to perform with basic regular expression
techniques
in a conversation analyzer 120 which may be any number of specific
implementations such
as a library in a programming language, or a separate application for text
processing, or
another common implementation of a regular expression system 120.
11
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
[049] Fig. 3 is a diagram showing two different approaches to assigning an
input embedded
word sequence vector 313 (shown here as a point projected onto a2 dimensional
plane
however typically the vectors have several hundred dimensions) to a single pre-
defined
category "C" of interest using a semantic distance "hypersphere" approach. The
black dots
Ai 311 and A2321, represent anchor points also in a 2-dimensional projection
310, 320 of
the high dimensional embedding vectors of the "anchor" text sequences used to
pre-define a
single category C. In the method on the left, the input embedded sequence
vector 313 is
assigned to the category C of interest if the distance from the input embedded
sequence
vector 313 to ANY anchor sequence vector 311, 321 is less than some threshold
tightness
.. distance "ri" or "r2" 312, 322 respectively for each anchor sequence vector
311, 321. In the
figure Fig. 3 the input embedded sequence vector 313 falls within a threshold
tightness
distance ri 312 from anchor sequence vector Ai 311 and so the input embedded
sequence
vector 313 is associated with category C, but this same input embedded
sequence vector 313
falls outside the distance "r2" 322 from the anchor sequence vector A2321 so
it is not
semantically close enough for A2321 to be the cause of it being associated
with the same
category C. If the input embedded sequence vector 313, falls outside both
distances ri 312
from Ai 311 and r2322 from A2321 then there is no semantic match made to any
anchor
sequence vector 311, 321 of category C and so the input embedded sequence
vector 313, is
not associated with category C. An alternate inferior approach is shown on the
right-hand
.. side 330 where the centroid 335 of in this example 4 anchor embedded
sequence vectors
331-334 for a category "D" is first computed and then the input embedded
sequence vector
340 is assigned to the category D only if it is less than a threshold distance
"r" 336 from the
centroid 335 of all 4 defining embedding sequences 331-334. The method on the
left is
preferred and performs particularly well when the anchor embedded sequence
vectors 313
.. and category are quite semantically heterogenous as described earlier. For
example if a
company sold products "apples", "oranges", "peaches" and "bananas", where in
this simple
case each anchor sequence 331-334 is only a single word, and these were used
as the anchor
sequences for the Product category, then the centroid 335 is also close to
each embedded
anchor vector 331-334 because they are all fruit and are semantically closed
so the results of
the approaches on the left and on the right would be similar. However if a
company's
products were quite semantically different, e.g. "apples", "shoes", "pencils"
and "computers"
then the centroid 335 of the embedded anchor vectors 331-334 is a blurred
semantic
"average" over the very different products and is not a good reference point
for assigning the
input embedded sequence vector 340 to the category.
12
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
[050] Fig. 4 is a block diagram of a system architecture for semantic
sentiment analysis,
according to a preferred aspect. Text input is provided 205, which may be
accomplished
using many methods to input text into a function or application as is common
in the art
including manual entry, web crawling, surveying users, and more, and such
input is loaded
into an input buffer 405. An input buffer 405 sends, once input data is loaded
into the buffer,
to both a final category and sentiment analysis engine 450 and a chunk parser
410. A chunk
parser 410 is capable of and responsible for partitioning provided buffered
input into chunks
determined by e.g. parts of speech (PoS) parsing and may be configured to
ignore certain
words, characters, or sentences based on content, according to a preferred
aspect. After
.. buffered 405 input 205 is parsed into chunks 410, chunks are sent to a
sentiment module
410a which computes sentiment at the chunk level and then passes it on to a
deterministic
rules engine 415 which uses deterministic rules to assign chunks to pre-
defined categories by
pattern matching, which it then stores 430. If the hard rules find no match in
the chunk, the
chunk is then sent to a semantic similarity engine 420 which then reduces the
chunk further
into a possibly reduced "sequence" of words (e.g. by selecting only the nouns
in the sequence
or up to and including the whole chunk) that it will embed into a semantic
vector space to
become an embedded sequence vector. Threshold semantic distance analysis with
a single
anchor point 310, 320 is then used to attempt to match the input embedded
sequence vector
to a pre-defined anchor sequence vector. If a match is found the semantic
match is stored in
.. 430. If no match is found, the input sequence embedding vector is then sent
from the
semantic similarity engine 420 to a semantic cluster discovery engine 425 to
determine
clusters of semantically similar input embedding vectors, which represents
further
categorization metadata and sentiment analysis opportunity for the system.
Data is examined
in the semantic cluster discovery engine 425 for non-predefined categories
which may be
procedurally generated by clustering to attempt to discover additional cluster
categories to
describe a given chunk of data. The semantic cluster discovery engine 425
stores the new or
additional cluster categories in the auto discovered category storage 440.
Category data from
pre-defined category storage 430 and auto-discovered category storage 440 for
a given chunk
of data is retrieved by the category and sentiment analysis engine 450, and
compared and
integrated together 445, to provide a comprehensive data profile of the
categories of
semantics, and such results are then stored in a combined category storage
unit 435. Buffered
input 405 and an integrated category 445 profile are both processed in a
category and
sentiment analysis engine 450, which processes the raw text and the metadata
about the
categories within different chunks of the data, and analyzes categories and
sentiments
13
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
expressed within the data, and displays the results of the analysis to the
user in the form of
text, graphics, or both, enabling for example, filtering by category, cross
tabbing, pivoting
and graphical chart views of the data including breakdowns by category,
sorting by sentiment
and showing trends over time.
[051] Fig. 5 is a block diagram of core components in a deterministic rules
engine 415,
according to a preferred aspect. Input from the compute sentiment at the chunk
level 411 is
fed into a deterministic rules engine 415. A regular expression categorizer
510 is responsible
for the actual pattern matching, and sends its output, after using an initial
regular expression
pattern search to find category matches, to both a semantic similarity engine
420 and a pre-
defined category storage 430 to store the results of the deterministic
category matching.
[052] Fig. 6 is a block diagram of core components in a semantic similarity
engine 420,
according to a preferred aspect. Input chunks are received from a
deterministic rules engine
415 and fed into a chunk to embedding sequence reducer 610 which reduces the
chunk
further into a possibly reduced "sequence" of words (e.g. by selecting only
the nouns in the
sequence or up to and including the whole chunk). The embedding sequence
preserves the
order of words from the original text and is different from a "bag of words"
approach where
word order is not important. For example "North America" has a distinct
meaning to simply
the presence of the words "America" and "North". The resulting embedding
sequence of
words is then sent to the sequence embedder 620, which will embed each input
word
sequence into a high dimensional vector according to the chosen sequence
embedding model
(e.g. Phrase2Vec as discussed above) which provides a numeric form of
measuring the
semantic meanings of a particular word sequence, to be used to determine the
category as
follows. A semantic distance comparator 630 determines the proximity of each
input
embedded sequence vector to the pre-defined Anchor Sequence vectors for each
potential
category match according to the method of Fig. 3(a) to find whether a semantic
match has
been made to a category or not. Output from the semantic similarity engine
420, specifically
from the semantic distance comparator 630,is then sent to a pre-defined
category storage
unit 430 and to a semantic cluster discovery engine 425 for unsupervised
categorization via
clustering.
[053] Fig. 7 is a block diagram of core components in a semantic cluster
discovery engine
425, according to a preferred aspect. Input is received from a semantic
similarity generator
420, and added to a storage area of previously unmatched embedded vectors 710.
The
semantic cluster identifier 720 finds possible clusters present in the
accumulated unmatched
14
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
vectors, before sending the data to a new category generator 730, for the
purpose of
generating new categories for possibly multi-semantic-anchor datasets, to
analyze clusters of
semantic data chunks.
[054] Fig. 8 is a block diagram of core components in a category and sentiment
analysis
engine 450, according to a preferred aspect. Input from category storage units
430, 435,
440, as well as from buffered but unprocessed input 405, is all joined
together in a full text
joiner 810, which joins all category IDs (pre-defined and discovered),
sentiment, text chunks
and embedding word sequences and metadata so far gathered and generated
preserving the
positions and range of each text item in the original text and also joins with
other structured
data attributes relating e.g. to the people (customer, customer segment,
employees, location,
channel, line of business etc.) involved in the conversation or text, plus
joining with other
available data using on the output category IDs. The joined dataset is then
passed to three
additional components. A cross tab analyzer 820 which enables an end-user to
manually
analyze, aggregate and sort tabular or graphical output by any of the
described attributes or
categories in the taxonomy, for example to show the conversations with extreme
high and low
sentiment concerning any category or level in the business taxonomy of
categories. Based on
the results the end-user of the system may want to "promote" a newly
discovered cluster to
become a regular -predefined category by causing a step to be taken that
stores the centroid
of the discovered cluster and its radius as a new pre-defined category with a
tightness radius
and having a convenient category name potentially given to it by the end-user.
The joined
data is also sent to the trend analyzer 830, which analyses and displays the
time dynamics of
the number of and proportion of texts in each category highlighting for the
user the growth or
decline of categories as well as optionally creating an automated management
alert which is
triggered when an emerging category grows rapidly and has for example a strong
negative
sentiment. The joined data is also sent to the supervised learner 840 (e.g. a
decision tree)
which takes all structured data attributes from the joiner including the
category ID and
sentiment level and learns the relation between attributes, so it can for
example predict which
combination of structured attributes is likely to results in texts that has
strongly negative
sentiment in a particular business category. In addition variable influence
scores from the
supervised output can highlight to the user which structured data attributes
are contribution
most to the likelihood of negative sentiment on a particular category.
[055] Note also that as described above a typical configuration is to follow
the procedure of
dividing the input text into noun phrase chunks, calculate the sentiment on
the chunks, then
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
reduce to only short noun sequences by the reducer 610 while performing the
semantic
embedding in the semantic similarity engine 420. By removing sentiment-laden
adjectives
before embedding we have effectively separated sentiment as an orthogonal
dimension by
which we can analyze the semantic categories. However, by the minor change of
allowing the
reducer 610 to do less filtering and also include adjectives and other emotion-
laden words we
can have categories that also include emotion. For example, "friendly
assistant" and
"unfriendly assistant" can be separate pre-defined or discovered categories
instead of having a
single pre-defined or discovered category "assistant" with cases where it has
positive
sentiment and others with negative sentiment. So, by changing the filtering on
the reducer
610 prior to embedding we can alter the balance between relying on the need to
have co-
occurrence of any kind of rare emotion-laden language near entities that were
present in the
text corpus used to create the embeddings model versus utilizing the language
co-occurrence
in the input string.
Detailed Description of Preferred Aspects
[056] Fig. 2 is a block diagram of a method of a system to analyze sentiment
in
conversations from users using new text analysis techniques, according to a
preferred
embodiment. Input text is provided 205 which may be transcriptions of natural
speech, or
surveys, or some other common source of possible natural language text 205.
This text is split
into chunks 210, using a chunk parser 410, which may split input text 205 into
chunks based
on parts of speech (e.g. noun phrases), word separators, sentence separation,
number of
characters, number of words, or some other desirable separation technique. The
sentiment is
then computed at the chunk level 215. Two different categorization modules
then attempt to
find a pre-defined category for the chunk. The first 220 uses simple rules and
pattern
matching (e.g. REGEX) which searches through a given chunk for any of the
matches 240,
including e.g. product names or brands 240a, people and personal names 240b,
process titles
and technical phrases 240c, or places and physical locations by name 240d. If
a match is
found the process stops and the chunk is assigned to the matched category. If
no match is
found the processing continues to categorization using semantic distance 225
is performed
using a chunk to embedding sequence reducer 610, a sequence embedder 620 and a
semantic
distance comparator 630. Auto discovery of other semantic clusters 230 occurs
next,
generating a plurality 245 of auto-generated categories 245a-245h using a new
category
generator 750, and both new 230 and pre-defined 220,225 categories are
analyzed along
16
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
with other factors in data chunks including a given perceived sentiment or a
time period 250.
Finally, focused action is taken 255, which involves extracting the final
category, context, and
sentiment values using a category and sentiment analysis engine 450, and
allowing analysts or
other users of the system to take action based on the results of the detailed
sentiment and
semantic analysis of the given data.
[057] Fig. 13 is a method diagram illustrating key high-level functionality
and steps taken in
the operation of a category assignment and sentiment analysis system,
according to a
preferred embodiment. Initial text data reception, parsing, and analysis 1310
is performed in
Fig. 2 by a variety of components including a deterministic rules engine 415.
Vectorization
of text 1320 is accomplished initially using a chunk-to-embedding sequence
reducer 610 in a
semantic similarity engine 420, which embeds a vector as metadata in a chunk,
or
alternatively this may be understood as generating additional data which is
kept separate
from the initial input text 405 and processed alongside it throughout the
system, as metadata
or "data about data." Categories are also determined after textual analysis
1330, through the
use of a sequence embedder 620 which embeds sequences of words or a sequence
of
characters into metadata for semantic threshold analysis, and a semantic
distance comparator
630 in a semantic similarity engine 420. These components draw correlations
between
categories and semantic vectors already determined about textual data, and
compose or
generate new categories and metadata about these relationships and a "bigger
picture" view
of the data, and do not have pre-determined or preset categories to choose
from, unlike the
operation of a deterministic rules engine 415. Categories may also be merged
or new
categories synthesized from two other categories identified in a chunk of
text, according to a
preferred aspect. Insertion of these categories and higher-level vectors 1340
is accomplished
using a semantic similarity engine 420, resulting in metadata that describes
trends and larger-
scope information about a given chunk, or this may be thought of as meta-
metadata. Further,
an analysis of the connections between all of the present vectors and
categories is performed
1350, using a semantic cluster discovery engine 425, especially using
subcomponents
including an unmatched vector storage unit 710, a semantic cluster identifier
720, and a new
category generator 730, which may be used to analyze the varying semantic
clusters and the
respective semantic "directions" they may pull a given chunk or complete set
of textual data,
or a time evolution analyzer 720 may be used to analyze a shift in categories
over the
timeline of a discussion input, or a sentiment analyzer 730 may be used to
analyze sentiment-
giving words or phrases such as "hate" or "love," or "enjoyable evening." A
semantic cluster
17
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
identifier 740 and new category generator 750 are further used in the task of
analyzing all
vectors and categories 1350, identifying clusters of semantically similar
phrases or chunk and
generating alternative new categories if needed to describe these semantic
clusters. Lastly,
using the vectors, categories, and clusters described previously, user
sentiment is determined
1360 using a category and sentiment analysis engine 450, parsing all text 810
and metadata
collected thus far to extract the categories of discussion 820, analyze the
context of pieces of
conversation 830, and determine user sentiment or "feeling" 840 based on all
collected data.
Hardware Architecture
[058] Generally, the techniques disclosed herein may be implemented on
hardware or a
.. combination of software and hardware. For example, they may be implemented
in an
operating system kernel, in a separate user process, in a library package
bound into network
applications, on a specially constructed machine, on an application-specific
integrated circuit
(ASIC), or on a network interface card.
[059] Software/hardware hybrid implementations of at least some of the aspects
disclosed
herein may be implemented on a programmable network-resident machine (which
should be
understood to include intermittently connected network-aware machines)
selectively activated
or reconfigured by a computer program stored in memory. Such network devices
may have
multiple network interfaces that may be configured or designed to utilize
different types of
network communication protocols. A general architecture for some of these
machines may be
.. described herein in order to illustrate one or more exemplary means by
which a given unit of
functionality may be implemented. According to specific aspects, at least some
of the features
or functionalities of the various aspects disclosed herein may be implemented
on one or more
general-purpose computers associated with one or more networks, such as for
example an
end-user computer system, a client computer, a network server or other server
system, a
.. mobile computing device (e.g., tablet computing device, mobile phone,
smartphone, laptop,
or other appropriate computing device), a consumer electronic device, a music
player, or any
other suitable electronic device, router, switch, or other suitable device, or
any combination
thereof. In at least some aspects, at least some of the features or
functionalities of the various
aspects disclosed herein may be implemented in one or more virtualized
computing
environments (e.g., network computing clouds, virtual machines hosted on one
or more
physical computing machines, or other appropriate virtual environments).
18
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
[060] Referring now to Fig. 9, there is shown a block diagram depicting an
exemplary
computing device 10 suitable for implementing at least a portion of the
features or
functionalities disclosed herein. Computing device 10 may be, for example, any
one of the
computing machines listed in the previous paragraph, or indeed any other
electronic device
capable of executing software- or hardware-based instructions according to one
or more
programs stored in memory. Computing device 10 may be configured to
communicate with
a plurality of other computing devices, such as clients or servers, over
communications
networks such as a wide area network a metropolitan area network, a local area
network, a
wireless network, the Internet, or any other network, using known protocols
for such
communication, whether wireless or wired.
[061] In one aspect, computing device 10 includes one or more central
processing units
(CPU) 12, one or more interfaces 15, and one or more busses 14 (such as a
peripheral
component interconnect (PCI) bus). When acting under the control of
appropriate software
or firmware, CPU 12 may be responsible for implementing specific functions
associated with
the functions of a specifically configured computing device or machine. For
example, in at
least one aspect, a computing device 10 may be configured or designed to
function as a server
system utilizing CPU 12, local memory 11 and/or remote memory 16, and
interface(s) 15. In
at least one aspect, CPU 12 may be caused to perform one or more of the
different types of
functions and/or operations under the control of software modules or
components, which for
example, may include an operating system and any appropriate applications
software,
drivers, and the like.
[062] CPU 12 may include one or more processors 13 such as, for example, a
processor
from one of the Intel, ARM, Qualcomm, and AMD families of microprocessors. In
some
aspects, processors 13 may include specially designed hardware such as
application-specific
integrated circuits (ASICs), electrically erasable programmable read-only
memories
(EEPROMs), field-programmable gate arrays (FPGAs), and so forth, for
controlling
operations of computing device 10. In a specific aspect, a local memory 11
(such as non-
volatile random access memory (RAM) and/or read-only memory (ROM), including
for
example one or more levels of cached memory) may also form part of CPU 12.
However,
there are many different ways in which memory may be coupled to system 10.
Memory 11
may be used for a variety of purposes such as, for example, caching and/or
storing data,
programming instructions, and the like. It should be further appreciated that
CPU 12 may be
one of a variety of system-on-a-chip (SOC) type hardware that may include
additional
19
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
hardware such as memory or graphics processing chips, such as a QUALCOMM
SNAPDRAGONTM or SAMSUNG EXYNOSTM CPU as are becoming increasingly
common in the art, such as for use in mobile devices or integrated devices.
[063] As used herein, the term "processor" is not limited merely to those
integrated circuits
referred to in the art as a processor, a mobile processor, or a
microprocessor, but broadly
refers to a microcontroller, a microcomputer, a programmable logic controller,
an
application-specific integrated circuit, and any other programmable circuit.
[064] In one aspect, interfaces 15 are provided as network interface cards
(NICs). Generally,
NICs control the sending and receiving of data packets over a computer
network; other types
of interfaces 15 may for example support other peripherals used with computing
device 10.
Among the interfaces that may be provided are Ethernet interfaces, frame relay
interfaces,
cable interfaces, DSL interfaces, token ring interfaces, graphics interfaces,
and the like. In
addition, various types of interfaces may be provided such as, for example,
universal serial bus
(USB), Serial, Ethernet, FIRE\VIRETM, THUNDERBOLTTm, PCI, parallel, radio
frequency
(RF), BLUETOOTHTm, near-field communications (e.g., using near-field
magnetics), 802.11
(WiFi), frame relay, TCP/IP, ISDN, fast Ethernet interfaces, Gigabit Ethernet
interfaces,
Serial ATA (SATA) or external SATA (ESATA) interfaces, high-definition
multimedia
interface (HDMI), digital visual interface (DVI), analog or digital audio
interfaces,
asynchronous transfer mode (ATM) interfaces, high-speed serial interface
(HSSI) interfaces,
Point of Sale (POS) interfaces, fiber data distributed interfaces (FDDIs), and
the like.
Generally, such interfaces 15 may include physical ports appropriate for
communication with
appropriate media. In some cases, they may also include an independent
processor (such as a
dedicated audio or video processor, as is common in the art for high-fidelity
A/V hardware
interfaces) and, in some instances, volatile and/or non-volatile memory (e.g.,
RAM).
[065] Although the system shown in Fig. 9 illustrates one specific
architecture for a
computing device 10 for implementing one or more of the aspects described
herein, it is by
no means the only device architecture on which at least a portion of the
features and
techniques described herein may be implemented. For example, architectures
having one or
any number of processors 13 may be used, and such processors 13 may be present
in a single
.. device or distributed among any number of devices. In one aspect, a single
processor 13
handles communications as well as routing computations, while in other aspects
a separate
dedicated communications processor may be provided. In various aspects,
different types of
features or functionalities may be implemented in a system according to the
aspect that
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
includes a client device (such as a tablet device or smartphone running client
software) and
server systems (such as a server system described in more detail below).
[066] Regardless of network device configuration, the system of the present
aspect may
employ one or more memories or memory modules (such as, for example, remote
memory
block 16 and local memory 11) configured to store data, program instructions
for the
general-purpose network operations, or other information relating to the
functionality of the
aspects described herein (or any combinations of the above). Program
instructions may
control execution of or comprise an operating system and/or one or more
applications, for
example. Memory 16 or memories 11, 16 may also be configured to store data
structures,
configuration data, encryption data, historical system operations information,
or any other
specific or generic non-program information described herein.
[067] Because such information and program instructions may be employed to
implement
one or more systems or methods described herein, at least some network device
aspects may
include nontransitory machine-readable storage media, which, for example, may
be
configured or designed to store program instructions, state information, and
the like for
performing various operations described herein. Examples of such nontransitory
machine-
readable storage media include, but are not limited to, magnetic media such as
hard disks,
floppy disks, and magnetic tape; optical media such as CD-ROM disks; magneto-
optical
media such as optical disks, and hardware devices that are specially
configured to store and
perform program instructions, such as read-only memory devices (ROM), flash
memory (as is
common in mobile devices and integrated systems), solid state drives (SSD) and
"hybrid SSD"
storage drives that may combine physical components of solid state and hard
disk drives in a
single hardware device (as are becoming increasingly common in the art with
regard to
personal computers), memristor memory, random access memory (RAM), and the
like. It
should be appreciated that such storage means may be integral and non-
removable (such as
RAM hardware modules that may be soldered onto a motherboard or otherwise
integrated
into an electronic device), or they may be removable such as swappable flash
memory
modules (such as "thumb drives" or other removable media designed for rapidly
exchanging
physical storage devices), "hot-swappable" hard disk drives or solid state
drives, removable
.. optical storage discs, or other such removable media, and that such
integral and removable
storage media may be utilized interchangeably. Examples of program
instructions include
both object code, such as may be produced by a compiler, machine code, such as
may be
produced by an assembler or a linker, byte code, such as may be generated by
for example a
21
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
JAVATM compiler and may be executed using a Java virtual machine or
equivalent, or files
containing higher level code that may be executed by the computer using an
interpreter (for
example, scripts written in Python, Perl, Ruby, Groovy, or any other scripting
language).
[068] In some aspects, systems according to the present aspect may be
implemented on a
standalone computing system. Referring now to Fig. 10, there is shown a block
diagram
depicting a typical exemplary architecture of one or more aspects or
components thereof on a
standalone computing system. Computing device 20 includes processors 21 that
may run
software that carry out one or more functions or applications of aspects, such
as for example a
client application 24. Processors 21 may carry out computing instructions
under control of an
operating system 22 such as, for example, a version of MICROSOFT WINDONVSTm
operating system, APPLE OSXTM or iOSTM operating systems, some variety of the
Linux
operating system, ANDROIDTM operating system, or the like. In many cases, one
or more
shared services 23 may be operable in system 20, and may be useful for
providing common
services to client applications 24. Services 23 may for example be \VINDO\VSTM
services,
user-space common services in a Linux environment, or any other type of common
service
architecture used with operating system 21. Input devices 28 may be of any
type suitable for
receiving user input, including for example a keyboard, touchscreen,
microphone (for
example, for voice input), mouse, touchpad, trackball, or any combination
thereof Output
devices 27 may be of any type suitable for providing output to one or more
users, whether
remote or local to system 20, and may include for example one or more screens
for visual
output, speakers, printers, or any combination thereof. Memory 25 may be
random-access
memory having any structure and architecture known in the art, for use by
processors 21, for
example to run software. Storage devices 26 may be any magnetic, optical,
mechanical,
memristor, or electrical storage device for storage of data in digital form
(such as those
described above, referring to Fig. 9). Examples of storage devices 26 include
flash memory,
magnetic hard drive, CD-ROM, and/or the like.
[069] In some aspects, systems of the present aspect may be implemented on a
distributed
computing network, such as one having any number of clients and/or servers.
Referring now
to Fig. 11, there is shown a block diagram depicting an exemplary architecture
30 for
implementing at least a portion of a system according to an aspect on a
distributed computing
network. According to the aspect, any number of clients 33 may be provided.
Each client 33
may run software for implementing client-side portions of the present aspect;
clients may
comprise a system 20 such as that illustrated in Fig. 10. In addition, any
number of servers
22
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
32 may be provided for handling requests received from one or more clients 33.
Clients 33
and servers 32 may communicate with one another via one or more electronic
networks 31,
which may be in various aspects any of the Internet, a wide area network, a
mobile telephony
network (such as CDMA or GSM cellular networks), a wireless network (such as
WiFi,
WiMAX, LTE, and so forth), or a local area network (or indeed any network
topology known
in the art. The aspect does not prefer any one network topology over any
other). Networks 31
may be implemented using any known network protocols, including for example
wired
and/or wireless protocols.
[070] In addition, in some aspects, servers 32 may call external services 37
when needed to
obtain additional information, or to refer to additional data concerning a
particular call.
Communications with external services 37 may take place, for example, via one
or more
networks 31. In various aspects, external services 37 may comprise web-enabled
services or
functionality related to or installed on the hardware device itself. For
example, in an aspect
where client applications 24 are implemented on a smartphone or other
electronic device,
client applications 24 may obtain information stored in a server system 32 in
the cloud or on
an external service 37 deployed on one or more of a particular enterprise's or
user's premises.
[071] In some aspects, clients 33 or servers 32 (or both) may make use of one
or more
specialized services or appliances that may be deployed locally or remotely
across one or
more networks 31. For example, one or more databases 34 may be used or
referred to by one
or more aspects. It should be understood by one having ordinary skill in the
art that databases
34 may be arranged in a wide variety of architectures and using a wide variety
of data access
and manipulation means. For example, in various aspects one or more databases
34 may
comprise a relational database system using a structured query language (SQL),
while others
may comprise an alternative data storage technology such as those referred to
in the art as
"NoSQL" (for example, HADOOP CASSANDRATM, GOOGLE BIGTABLETm, and so
forth). In some aspects, variant database architectures such as column-
oriented databases, in-
memory databases, clustered databases, distributed databases, or even flat
file data
repositories may be used according to the aspect. It will be appreciated by
one having
ordinary skill in the art that any combination of known or future database
technologies may
be used as appropriate, unless a specific database technology or a specific
arrangement of
components is specified for a particular aspect herein. Moreover, it should be
appreciated
that the term "database" as used herein may refer to a physical database
machine, a cluster of
machines acting as a single database system, or a logical database within an
overall database
23
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
management system. Unless a specific meaning is specified for a given use of
the term
"database", it should be construed to mean any of these senses of the word,
all of which are
understood as a plain meaning of the term "database" by those having ordinary
skill in the
art.
.. [072] Similarly, most aspects may make use of one or more security systems
36 and
configuration systems 35. Security and configuration management are common
information
technology (IT) and web functions, and some amount of each are generally
associated with
any IT or web systems. It should be understood by one having ordinary skill in
the art that
any configuration or security subsystems known in the art now or in the future
may be used in
conjunction with aspects without limitation, unless a specific security 36 or
configuration
system 35 or approach is specifically required by the description of any
specific aspect.
[073] Fig. 12 shows an exemplary overview of a computer system 40 as may be
used in any
of the various locations throughout the system. It is exemplary of any
computer that may
execute code to process data. Various modifications and changes may be made to
computer
system 40 without departing from the broader scope of the system and method
disclosed
herein. Central processor unit (CPU) 41 is connected to bus 42, to which bus
is also
connected memory 43, nonvolatile memory 44, display 47, input/output (I/O)
unit 48, and
network interface card (NIC) 53. I/O unit 48 may, typically, be connected to
keyboard 49,
pointing device 50, hard disk 52, and real-time clock 51. NIC 53 connects to
network 54,
which may be the Internet or a local network, which local network may or may
not have
connections to the Internet. Also shown as part of system 40 is power supply
unit 45
connected, in this example, to a main alternating current (AC) supply 46. Not
shown are
batteries that could be present, and many other devices and modifications that
are well
known but are not applicable to the specific novel functions of the current
system and method
disclosed herein. It should be appreciated that some or all components
illustrated may be
combined, such as in various integrated applications, for example Qualcomm or
Samsung
system-on-a-chip (SOC) devices, or whenever it may be appropriate to combine
multiple
capabilities or functions into a single hardware device (for instance, in
mobile devices such as
smartphones, video game consoles, in-vehicle computer systems such as
navigation or
.. multimedia systems in automobiles, or other integrated hardware devices).
[074] In various aspects, functionality for implementing systems or methods of
the various
aspects described herein may be distributed among any number of client and/or
server
components. For example, various software modules may be implemented for
performing
24
CA 03131157 2021-08-20
WO 2020/172649
PCT/US2020/019438
various functions in connection with the aspect, and such modules may be
variously
implemented to run on server and/or client components.
[075] The skilled person will be aware of a range of possible modifications of
the various
aspects described above. Accordingly, the present invention is defined by the
claims and their
equivalents.
