# US 2019/0377748 A1 · Method and system for sentiment analysis of information > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0377748 A1 - Document ID: US20190377748A1 - Status: Published application - Kind: A1 - Application number: 16/550,479 - Filed: 2019-08-26 - Published: 2019-12-12 - Earliest claimed: 2018-02-12 - Assignees: Hithink Royalflush Information Network Co., Ltd. - Applicants: Hithink Royalflush Information Network Co., Ltd. - Inventors: Zheng YI; Wei XIA - Web page: https://patentgrep.com/patent/US20190377748A1 ## CPC - G06F 16/35: Clustering; Classification - G06F 16/958: Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking - G06F 18/22: Matching criteria, e.g. proximity measures - G06F 18/2411: based on the proximity to a decision surface, e.g. support vector machines - G06F 18/2413: based on distances to training or reference patterns - G06F 40/242: Dictionaries - G06F 40/30: Semantic analysis - G06V 10/75: Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries - G06V 10/764: using classification, e.g. of video objects - G06V 30/373: {using a special pattern or subpattern alphabet} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract One aspect of the present disclosure relates to a method of sentiment analysis based on ambiguity analysis, which includes analyzing information with the sentiment analysis models and the ambiguity analysis models. Another aspect of the present disclosure relates to a method of training the sentiment analysis models and ambiguity analysis models, which includes acquiring information, constructing lexicons, conducting sentiment analysis and ambiguity analysis with said lexicons, acquiring corpus, and training models, etc. Meanwhile, another aspect of the present disclosure relates to a system of sentiment analysis, which includes input, and output modules, acquisition modules, processing modules and database. ## Claims ### Claim 1 A method implemented on at least one device each of which has at least one processor and a storage medium, the method comprising: obtaining a user input; retrieving sentimental information from a database according to the user input, each sentimental information marked with a sentiment type; and classifying the sentimental information based on the sentiment types, wherein the sentiment type is marked by the following steps: acquiring preliminary information from the database according to the user input; constructing a repository by storing the preliminary information in the repository; retrieving a sentiment lexicon by retrieving sentimental words or sentimental collocations from the preliminary information and storing the sentimental words or sentimental collocations in the sentiment lexicon; matching the preliminary information with the sentiment lexicon to obtain a set of sentimental sentences; and analyzing the set of sentimental sentences and marking with corresponding sentiment types. ### Claim 2 (depends on 1) The method of claim 1, further comprising: performing an ambiguity analysis on the preliminary information to determine non-ambiguous information or ambiguous information. ### Claim 3 (depends on 2) The method of claim 2, wherein the performing the ambiguity analysis further comprises: determining a keyword lexicon by retrieving terminologies from the preliminary information and storing the terminologies in the keyword lexicon; matching the preliminary information in the repository with terminologies in the keyword lexicon to obtain matching results; scoring the matching results to obtain scores based on the terminologies in the keyword lexicon; and identifying the non-ambiguous information or the ambiguous information from the preliminary information according to the scores. ### Claim 4 (depends on 3) The method of claim 3, further comprising: analyzing ambiguity of the terminologies in the keyword lexicon to construct an ambiguity list, the ambiguity list storing terms of art that are verified ambiguous; and scoring the matching results to obtain scores further based on ambiguity list. ### Claim 5 (depends on 3) The method of claim 3, wherein the identifying the non-ambiguous information or the ambiguous information from the preliminary information according to the scores further comprises: determining an information as ambiguous information if the score of the information is no more than a first threshold; determining an information as non-ambiguous information if the score of the information is no less than a second threshold. ### Claim 6 (depends on 5) The method of claim 5, further comprising: determining an information as non-ambiguous information or ambiguous information by an operator or by ambiguity analysis model if the score of the information is more than the first threshold and less than the second threshold. ### Claim 7 (depends on 6) The method of claim 6, wherein the ambiguity analysis model comprises a Maximum entropy algorithm training model. ### Claim 8 (depends on 5) The method of claim 5, wherein the first threshold or the second threshold are fixed. ### Claim 9 (depends on 5) The method of claim 5, wherein the first threshold or the second threshold are adjustable based on an amount of the preliminary information. ### Claim 10 (depends on 1) The method of claim 1, wherein the sentiment type is further determined using a sentiment analyzer. ### Claim 11 (depends on 10) The method of claim 10, wherein the sentiment analyzer comprises at least one of a Maximum entropy algorithm training model or a support vector machine (SVM) model. ### Claim 12 (depends on 1) The method of claim 1, further comprising: obtaining a second user input; obtaining second sentimental information marked with sentiment types corresponding to the second user input; displaying a comparison of a number of sentimental information marked with the same sentiment type between the user input and the second user input appears in predetermined time period. ### Claim 13 A system, comprising: a storage device storing a set of instructions; and at least one processor in communication with the storage device, wherein when executing the set of instructions, the at least one processor is configured to: obtain a user input; retrieve sentimental information from a database according to the user input, each sentimental information marked with a sentiment type; and classify the sentimental information based on the sentiment types, wherein to mark the sentiment type, the at least one processor is configured to: acquire preliminary information from the database according to the user input; construct a repository by storing the preliminary information in the repository; retrieve a sentiment lexicon by retrieving sentimental words or sentimental collocations from the preliminary information and storing the sentimental words or sentimental collocations in the sentiment lexicon; match the preliminary information with the sentiment lexicon to obtain a set of sentimental sentences; and analyze the set of sentimental sentences and marking with corresponding sentiment types. ### Claim 14 (depends on 13) The system of claim 13, the at least one processor is configured to: perform an ambiguity analysis on the preliminary information to determine non-ambiguous information or ambiguous information. ### Claim 15 A method implemented on at least one device each of which has at least one processor and a storage medium, the method comprising: acquiring information; constructing a repository by storing the information in the repository; retrieving a sentiment lexicon by retrieving sentimental collocations from the information and storing the sentimental collocations in the sentiment lexicon; matching the information with the sentiment lexicon to obtain a set of sentimental sentences; analyzing the set of sentimental sentences and marking with corresponding sentiment types; retrieving a corpus including the sentimental collocations marked with the sentiment types based on the set of sentimental sentences marked with the sentiment types; and training a sentiment analyzer using the corpus. ### Claim 16 (depends on 15) The method of claim 15, wherein the information includes non-ambiguous information or ambiguous information after an ambiguity analysis. ### Claim 17 (depends on 15) The method of claim 15, wherein the retrieving the corpus further comprising determining a classification accuracy for each sentimental collocation based on a number of sentimental sentences marked with a sentiment type matched by the sentimental collocation and a number of sentimental sentences matched by the sentimental collocation; determining the sentimental collocation as a strong sentimental collocation marked with the sentiment type if the classification accuracy is more than a default threshold; and storing the strong sentimental collocation marked with the sentiment type to the corpus. ### Claim 18 (depends on 15) The method of claim 15, wherein the sentiment analyzer comprises at least one of a Maximum entropy algorithm training model or a support vector machine (SVM) model. ### Claim 19 (depends on 15) The method of claim 15, further comprising analyzing the set of sentimental sentences and marking with corresponding sentiment types by an operator. ### Claim 20 (depends on 15) The method of claim 15, wherein the sentiment analyzer is updated periodically.