System and method for text categorization and sentiment analysis

1. A system for categorization and sentiment analysis, 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 to generate an embedded sequence vector in a semantic vector space from a sequence of words, wherein the chunks of text comprising words and phrases; and

compute sentiment on the text at the chunk level, using a chunk parser; 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 the chunked 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 tire 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 draw correlations between the categories and semantic vectors and generate metadata for the correlation; 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 including writer's attitude, emotion, ideological, quantitative analysis;

analyze the sentiment of categories of interest to a user of the system, wherein the analyzing includes determining user sentiment using the vectors, categories, and clusters; and

output the results of the analysis to tire user in the form of text, graphics, or both, wherein the results are sorted by sentiment and showing trends over the time.

2. A method for categorization and sentiment analysis, comprising the steps of:

receiving input in text form;

breaking the text into chunks of text to generate an embedded sequence vector in a semantic vector space from a sequence of words, wherein the 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 associations, using a semantic cluster discovers engine and draw correlations between the categories and semantic vectors and generate metadata for the correlation;

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, where sentiment analysis includes quantitative analysis of one or more of writer's attitude, emotion, or ideology;

analyzing the sentiment of categories of interest to a user of the system, using a category and sentiment analysis engine, wherein the analyzing includes determining user sentiment using the vectors, categories, and clusters; 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, wherein the results are sorted by sentiment and showing trends over the time.