# US 2018/0322115 A1 · Method and device for determining comment, server and storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0322115 A1 - Document ID: US20180322115A1 - Status: Published application - Kind: A1 - Application number: 15/854,918 - Filed: 2017-12-27 - Published: 2018-11-08 - Earliest claimed: 2017-05-08 - Assignees: Beijing Baidu Netcom Science Technology Co., Ltd. - Applicants: Beijing Baidu Netcom Science and Technology Co., Ltd. - Inventors: Yingchao SHI; Wei He; Qiaoqiao She; Jianqing Cui; Xiangyang Zhou; Junqiang Zheng - Web page: https://patentgrep.com/patent/US20180322115A1 ## CPC - G06F 40/30: Semantic analysis - G06F 16/951: Indexing; Web crawling techniques - G06F 40/117: Tagging; Marking up (details of markup languages G06F40/143); Designating a block; Setting of attributes (style sheets, e.g. eXtensible Stylesheet Language Transformation [XSLT], G06F40/154) - G06F 40/169: Annotation, e.g. comment data or footnotes - G06F 40/247: Thesauruses; Synonyms - G06F 40/279: Recognition of textual entities - G06F 40/295: Named entity recognition - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - G06N 5/022: Knowledge engineering; Knowledge acquisition - G06N 20/00: Machine learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Embodiments of the present disclosure provide a method and a device for determining a comment, a server and a storage medium. The method includes: obtaining text content of an event; determining an event tag of the event according to the text content; comparing the event tag with comment tags in a comment graph, to determine candidate comments, the comment graph being generated according to other events and corresponding comments, and the comment graph including comments and corresponding comment tags; and screening the candidate comments, to determine a candidate comment satisfying a preset condition as the comment of the event. ## Claims ### Claim 1 A method for determining a comment, comprising: obtaining text content of an event; determining an event tag of the event according to the text content; comparing the event tag with comment tags in a comment graph, to determine candidate comments, the comment graph being generated according to other events and corresponding comments, and the comment graph comprising comments and corresponding comment tags; and screening the candidate comments, to determine a candidate comment satisfying a preset condition as the comment of the event. ### Claim 2 (depends on 1) The method according to claim 1, wherein determining the event tag of the event according to the text content comprises: pre-processing the text content, to obtain a word set corresponding to the text content, the word set comprising at least two words; determining weight values of words in the word set according to a preset learning model; and determining a word corresponding to a weight value satisfying a condition as the event tag of the event. ### Claim 3 (depends on 2) The method according to claim 2, before pre-processing the text content, further comprising: extracting text elements in the text content, the text element comprising at least two key words; accordingly, after determining the word corresponding to the weight value satisfying the condition as the event tag of the event, further comprising: determining other key words comprising the event tag in the text elements as the event tag of the event. ### Claim 4 (depends on 1) The method according to claim 1, wherein comparing the event tag with the comment tags in the comment graph, to determine the candidate comments comprises: comparing the event tag with the comment tags corresponding to the comments in the comment graph, and sorting the comments from high to low according to the number of the event tags hit by the comment tags, to determine comments that rank high as the candidate comments. ### Claim 5 (depends on 4) The method according to claim 4, wherein screening the candidate comments, to determine the candidate comment satisfying the preset condition as the comment of the event comprises: determining a candidate comment that the comment tag is consistent and positive with the event tag of the event as the comment of the event. ### Claim 6 (depends on 4) The method according to claim 4, wherein, the event comprises a news, accordingly, after determining the candidate comment satisfying the preset condition as the comment of the event, the method further comprises: determining an emotional logic of the news, and commenting on the news by selecting qualified comment based on the emotional logic. ### Claim 7 (depends on 1) The method according to claim 1, wherein a process of generating the comment graph by other events and corresponding comments comprises: crawling and determining the comment through a web crawler; recognizing the comment through a preset recognition model, to determine a comment main body and a comment viewpoint corresponding to the comment; determining the comment tag corresponding to the comment according to the comment main body, the comment viewpoint and the event corresponding to the comment; and storing the comment and the determined comment tag as the comment graph in a structured form. ### Claim 8 A server, comprising: one or more processors; a storage device, configured to store one or more programs; wherein when the one or more programs are executed by the one or more processors, the one or more processors are configured to be caused to implement acts of: obtaining text content of an event; determining an event tag of the event according to the text content; comparing the event tag with comment tags in a comment graph, to determine candidate comments, the comment graph being generated according to other events and corresponding comments, and the comment graph comprising comments and corresponding comment tags; and screening the candidate comments, to determine a candidate comment satisfying a preset condition as the comment of the event. ### Claim 9 (depends on 8) The server according to claim 8, wherein the one or more processors are configured to determine the event tag of the event according to the text content by acts of: pre-processing the text content, to obtain a word set corresponding to the text content, the word set comprising at least two words; determining weight values of words in the word set according to a preset learning model; and determining a word corresponding to a weight value satisfying a condition as the event tag of the event. ### Claim 10 (depends on 9) The server according to claim 9, wherein the one or more processors are further configured to implement acts of: before pre-processing the text content, extracting text elements in the text content, the text element comprising at least two key words; accordingly, after determining the word corresponding to the weight value satisfying the condition as the event tag of the event, determining other key words comprising the event tag in the text elements as the event tag of the event. ### Claim 11 (depends on 8) The server according to claim 8, wherein the one or more processors are configured to compare the event tag with the comment tags in the comment graph, to determine the candidate comments by acts of: comparing the event tag with the comment tags corresponding to the comments in the comment graph, and sorting the comments from high to low according to the number of the event tags hit by the comment tags, to determine comments that rank high as the candidate comments. ### Claim 12 (depends on 11) The server according to claim 11, wherein the one or more processors are configured to screen the candidate comments, to determine the candidate comment satisfying the preset condition as the comment of the event by acts of: determining a candidate comment that the comment tag is consistent and positive with the event tag of the event as the comment of the event. ### Claim 13 (depends on 11) The server according to claim 11, wherein the event comprises a news, accordingly, the one or more processors are further configured to implement acts of: after determining the candidate comment satisfying the preset condition as the comment of the event, determining an emotional logic of the news, and commenting on the news by selecting qualified comment based on the emotional logic. ### Claim 14 (depends on 8) The server according to claim 8, wherein the one or more processors are further configured to implement acts of: crawling and determining the comment through a web crawler; recognizing the comment through a preset recognition model, to determine a comment main body and a comment viewpoint corresponding to the comment; determining the comment tag corresponding to the comment according to the comment main body, the comment viewpoint and the event corresponding to the comment; and storing the comment and the determined comment tag as the comment graph in a structured form. ### Claim 15 A non-transitory computer readable storage medium comprising computer executable instructions, wherein when the computer executable instructions are executed by a computer processor, a method for determining comment is executed, the method comprises: obtaining text content of an event; determining an event tag of the event according to the text content; comparing the event tag with comment tags in a comment graph, to determine candidate comments, the comment graph being generated according to other events and corresponding comments, and the comment graph comprising comments and corresponding comment tags; and screening the candidate comments, to determine a candidate comment satisfying a preset condition as the comment of the event. ### Claim 16 (depends on 15) The non-transitory computer readable storage medium according to claim 15, wherein determining the event tag of the event according to the text content comprises: pre-processing the text content, to obtain a word set corresponding to the text content, the word set comprising at least two words; determining weight values of words in the word set according to a preset learning model; and determining a word corresponding to a weight value satisfying a condition as the event tag of the event. ### Claim 17 (depends on 16) The non-transitory computer readable storage medium according to claim 16, before pre-processing the text content, the method further comprises: extracting text elements in the text content, the text element comprising at least two key words; accordingly, after determining the word corresponding to the weight value satisfying the condition as the event tag of the event, further comprising: determining other key words comprising the event tag in the text elements as the event tag of the event. ### Claim 18 (depends on 15) The non-transitory computer readable storage medium according to claim 15, wherein comparing the event tag with the comment tags in the comment graph, to determine the candidate comments comprises: comparing the event tag with the comment tags corresponding to the comments in the comment graph, and sorting the comments from high to low according to the number of the event tags hit by the comment tags, to determine comments that rank high as the candidate comments. ### Claim 19 (depends on 18) The non-transitory computer readable storage medium according to claim 18, wherein screening the candidate comments, to determine the candidate comment satisfying the preset condition as the comment of the event comprises: determining a candidate comment that the comment tag is consistent and positive with the event tag of the event as the comment of the event. ### Claim 20 (depends on 18) The non-transitory computer readable storage medium according to claim 18, wherein, the event comprises a news, accordingly, after determining the candidate comment satisfying the preset condition as the comment of the event, the method further comprises: determining an emotional logic of the news, and commenting on the news by selecting qualified comment based on the emotional logic.