CN110852095A - Statement hot spot extraction method and system - Google Patents
Statement hot spot extraction method and system Download PDFInfo
- Publication number
- CN110852095A CN110852095A CN201810871010.0A CN201810871010A CN110852095A CN 110852095 A CN110852095 A CN 110852095A CN 201810871010 A CN201810871010 A CN 201810871010A CN 110852095 A CN110852095 A CN 110852095A
- Authority
- CN
- China
- Prior art keywords
- syntax tree
- dependency
- emotion
- keyword
- sentence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Machine Translation (AREA)
Abstract
The invention relates to a statement hot spot extraction method, which comprises the following steps: extracting at least one emotion keyword from the document; wherein the document comprises at least one statement; performing dependency syntax analysis on the first statement to generate a dependency syntax tree; pruning the dependency syntax tree based on the first emotion keyword to form a result syntax tree; wherein the first emotion keyword is derived from a first sentence; forming a text vector set based on the result syntax tree; and clustering the set of text vectors to form at least one class of interest. The method is beneficial to improving the accuracy of the clustering result, does not influence the original semantics, and can more accurately express the hot points concerned by the client.
Description
Technical Field
The invention relates to the technical field of text analysis, in particular to a sentence hotspot extracting method and system.
Background
The customer service center can receive a large amount of customer complaints and consultations, then classifies the consultative documents according to the actual consultation categories of the customers through relevant service personnel such as telephone operators and the like, imports the results into the database, then obtains consultation hotspots according to the statistical information of the consultation documents, and analyzes the obtained consultation hotspots.
However, as the consulting contents of clients are increasingly diversified and the number of consulting documents is increased, a large number of consulting documents are classified only by a manual mode, and then statistics is carried out on the classified documents to obtain consulting hotspots and hotspot analysis is carried out, so that time and labor are consumed, and the method is too dependent on the subjectivity of analysts. If the sentences in the consultation document can be extracted and automatically analyzed, the evaluation and feedback of the client can be obtained in time, and the product or service can be updated/upgraded conveniently.
The prior art provides the following technical solutions: the method comprises the steps of extracting k consultation documents from a plurality of consultation documents, using the k consultation documents as k document categories, namely as the centers of initial clustering, calculating the similarity between other consultation documents and each document category respectively, classifying each consultation document into the document category with the highest similarity, further realizing the automatic classification process, and then extracting keywords to obtain the statistical information of one type of consultation documents or generating corresponding consultation hotspots.
However, the information involved in the conversation between the client and the customer service staff is very rich, many pieces of information are not related to the problem that the client wants to consult, and the method proposed by the prior art is not suitable for filtering irrelevant noise, and the noise may cover effective data, so that the document clustering process is inaccurate, the clustering algorithm execution efficiency is low, and the expected analysis effect is greatly different from the actual effect.
Disclosure of Invention
The invention aims to provide a statement hot spot extraction method which can more effectively filter noise and uninteresting information and improve clustering accuracy and clustering algorithm execution efficiency.
In order to achieve the above object, the present invention provides a technical solution as follows.
A sentence hot spot extraction method comprises the following steps: a) extracting at least one emotion keyword from the document; wherein the document comprises at least one statement; b) performing dependency syntax analysis on the first statement to generate a dependency syntax tree; c) pruning the dependency syntax tree based on the first emotion keyword to form a result syntax tree; wherein the first emotion keyword is derived from a first sentence; d) forming a text vector set based on the result syntax tree; and e) clustering the set of text vectors to form at least one class of interest.
Preferably, the method further comprises: for each class of interest: extracting at least one hotspot keyword; generating at least one hot spot expression based on the word frequency information of each hot spot keyword, wherein the hot spot expression corresponds to a permutation and combination of the at least one hot spot keyword; and determining hot spot expressions based on the support degree of each hot spot description in the document.
Preferably, step c) comprises: c1) determining a root node of the dependency syntax tree; c2) and performing breadth-first traversal based on the root node, and cutting out the nodes and the subtrees of the nodes of which the grammatical relations are not in accordance with the dependency relationship white list from the dependency syntax tree.
Preferably, the root node is determined in step c1) by: and taking the emotion keywords as the current search leaf nodes, searching upwards to the root nodes along the dependency syntax tree, or determining the nodes of which the current parallel relation reaches the maximum depth limit value as the root nodes.
Preferably, the dependency white list comprises: COO; a VOB; ADV; and, POB.
Preferably, in step b): the grammatical relation and the part of speech of each node of the dependency syntax tree respectively meet the LTP specification and the ICTCCLAS standard.
Preferably, step a) specifically comprises: sentences that do not contain any emotion keywords are excluded from the document.
Preferably, the clustering algorithm adopted in step e) comprises: and (4) a K-means clustering algorithm.
The invention also discloses a statement hot spot extraction system, which comprises: the emotion keyword extraction unit is used for extracting at least one emotion keyword from the document; a dependency syntax tree generating unit for performing dependency syntax analysis on the first sentence to generate a dependency syntax tree; wherein the first sentence is derived from a document; a result syntax tree generating unit coupled to the emotion keyword extracting unit and the dependency syntax tree generating unit, respectively, for pruning the dependency syntax tree based on the first emotion keyword to form a result syntax tree; wherein the first emotion keyword is derived from a first sentence; a vector conversion unit for forming a set of text vectors based on the result syntax tree; and a vector clustering unit for clustering the text vector set to form at least one interest class.
Before clustering analysis is carried out, the sentence hotspot extracting method adopts a mode of pruning the dependency syntax tree based on the emotion keywords to refine the client consulting document, so that the accuracy of a clustering result is improved, the method does not influence the original semantics, and can accurately express hotspots concerned by the client. In addition, when the hot spot expression is formed, the method can synthesize the expression sentences by considering the word frequency, so that the automatically extracted consultation hot spot is representative and easy to understand.
Drawings
Fig. 1 shows a flowchart of a statement hot spot extraction method according to a first embodiment of the present invention.
Fig. 2 is a schematic block diagram illustrating a sentence hot-spot lifting system according to a second embodiment of the present invention.
Detailed Description
In the following description specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art that embodiments of the invention may be practiced without these specific details. In the present invention, specific numerical references such as "first element", "second device", and the like may be made. However, specific numerical references should not be construed as necessarily subject to their literal order, but rather construed as "first element" as opposed to "second element".
The specific details set forth herein are merely exemplary and may be varied while remaining within the spirit and scope of the invention. The term "coupled" is defined to mean either directly connected to a component or indirectly connected to the component via another component.
Preferred embodiments of methods, systems and devices suitable for implementing the present invention are described below with reference to the accompanying drawings. Although embodiments are described with respect to a single combination of elements, it is to be understood that the invention includes all possible combinations of the disclosed elements. Thus, if one embodiment includes elements A, B and C, while a second embodiment includes elements B and D, the invention should also be considered to include A, B, C or the other remaining combinations of D, even if not explicitly disclosed.
As shown in fig. 1, a first embodiment of the present invention provides a statement hot spot extraction method, which includes the following steps.
And step S10, extracting at least one emotion keyword from the document.
The consultation document comprises a dialogue record of a client and the customer service center in a period of time or about a certain problem, wherein a plurality of sentences are recorded, and the client sentences and the customer service sentences are protected.
As an example, a consulting document records a query dialog of a client and a customer service with respect to transaction records, as shown in the following table.
First, the collected customer service dialogue data may be preprocessed to remove the meaningless information or noise such as "xxx number customer service person saying", "visitor xxx saying" and so on, so as to obtain the original corpus, whose content is shown in the following table.
Subsequently, a set of emotion keywords is determined from the original corpus. The inventor has found statistically that in a consultation session between a client and a customer service (visitor), words (e.g. words with emotional colors) are partially added
"not", "none", "no", etc.) are closely related to the hotspots of interest to the client, and thus, these keywords can be determined as emotion keywords.
According to step S10, at least one emotion keyword can be extracted from the counsel document. The black boxes in the table above indicate that the words contain the emotion keyword "none" and thus can be used for subsequent analysis. Preferably, sentences not containing any emotion keywords can be excluded from the original corpus, so that the accuracy of subsequent clustering and the execution efficiency of the algorithm are improved.
Step S12, the dependency syntax analysis is performed on the first sentence, and a dependency syntax tree is generated.
Wherein the first sentence is derived from an original corpus, and the original corpus is a result of preprocessing the consulting document. The statement "why I did not get the immune offer" shown in the black box in the table above is an example of the first statement.
Because the emotion keywords can reflect the hot spots focused by the client, according to the embodiment of the present invention, the sentences containing the emotion keywords in the original corpus can be respectively used as the "first sentences" in the present invention, which is a preferred embodiment of the present invention. In other words, steps S12 and S14 may be performed for each sentence in the original corpus that includes emotion keywords.
As an alternative embodiment, in the case that there are a plurality of emotion keywords extracted from the original corpus and they can be ranked according to priority, one sentence containing the highest priority emotion keyword is extracted from the original corpus as the first sentence, or sentences containing more emotion keywords are taken as the first sentence, which does not need to analyze all sentences containing emotion keywords in the original corpus, so that the efficiency of sentence analysis can be improved.
Dependency Parsing (DP) is to analyze the Dependency relationship between components in a language unit to reveal the syntactic structure. Intuitively, the dependency grammar analysis identifies grammar components of "principal object" and "fixed shape complement" in the sentence, and analyzes the relationship between the components. As an example, the client in the consulting document says "why I did not get the exemption offer", and its dependency parsing results can generate a dependency syntax tree based on the dependency parsing results as follows.
Preferably, in the process of generating the dependency syntax tree based on the dependency syntax analysis, the part-of-speech tag of each node of the dependency syntax tree conforms to the ICTCLAS standard, and the syntactic relation between each node satisfies the LTP specification. Alternatively, the parsing may be performed using the Stanford Parser work.
Part-of-speech Tagging (POS) is the task of giving each word in a sentence a Part-of-speech category. The part-of-speech category here may be nouns, verbs, adjectives or others. Specifically, v represents a verb, n represents a noun, r represents a pronoun, c represents a conjunct, a represents an adjective, d represents an adverb, and wp represents a punctuation mark. According to the LTP specification, the grammatical relations between words include various SBV predicate relations, VOB move-guest relations, IOB mediate relations, FOB pre-object, DBL bilingual, ATT centering relations, ADV middle relations, CMP move-complement relations, COO parallel relations, POB mediate relations, and the like.
And step S14, pruning the dependency syntax tree based on the first emotion keyword to form a result syntax tree.
This step is one of the key steps of the present invention, according to which the first emotion keyword is derived from the first sentence. As an example, the first sentence is "why i did not get the exemption offer", and the corresponding first emotion keyword is the word "none" in the first sentence.
Specifically, this step S14 may be performed as follows. First, the root node of each dependent syntax tree is determined. Secondly, breadth-first traversal is performed from the root node, and nodes and corresponding subtrees of which the grammatical relations do not conform to the dependency white list are pruned from the dependency syntax tree.
As an example, with the emotion keyword as the current search leaf node, the Root node is searched along the dependency syntax tree until the Root node (Root) is searched or the current side-by-side relationship ("COO") reaches the maximum depth limit value (e.g., set to 2), the node is determined as the syntax tree Root node, and the node is added to the resulting syntax tree. For example, the first sentence is "I why no exemption offer was obtained", its emotion keyword is "none", its parent node is
The parent node of "get" and "get" is "Root", and thus it is determined that "get" is the Root node of the current syntax tree.
Then, starting from the obtained dependency syntax tree root node, a restrictive breadth-first traversal is performed. In one implementation, only those nodes and corresponding sub-trees whose grammatical relationships between parent and child nodes do not fit into the dependency white list are pruned from the syntax tree in a breadth-first traversal. As another implementation, those nodes whose grammatical relations and parts of speech between the parent node and the child node satisfy the dependency white list are retained, and other nodes are all pruned from the syntax tree together with the corresponding subtree until traversal is completed for all leaf nodes or the current parallel relation ("COO") reaches the maximum depth.
Preferably, the dependency white list is shown in the following table, wherein only a part of the preferred white list is shown, e.g. the grammatical relations COO, VOB, ADV and POB. It will be appreciated that the white list may be modified depending on the particular application scenario.
For example, in the first sentence "i why no exemption offer was obtained," obtain "
The child nodes (with the part of speech v) are 'why' (with the part of speech r), 'me' (with the part of speech r)
"none" (part of speech d) and "exempt" (part of speech v); "get" and "why" relationship is ADV, but part of speech does not fit into the dependency white list, so "why" is discarded; the "none" relationship is that "ADV" is white-listed in dependency, so "none" is added to the result syntax tree, and similarly, "get" is also added to the result syntax tree, and "i" is also discarded.
In this way, pruning of the dependency syntax tree is completed, and the resulting syntax tree corresponds to the statement "no exemption offer is obtained", and the syntax structure is significantly simplified relative to the dependency syntax tree. In general, from each first sentence containing an emotion keyword, a corresponding resulting syntax tree can be derived; it is also possible that the dependency syntax tree of the first statement is pruned in its entirety during the pruning process.
Step S16, forming a text vector set based on the resulting syntactic tree.
In this step, a text vector set can be formed from at least one dependency syntax tree using one _ hot, tf-idf, wordvec, etc. text-to-digital techniques, wherein the text vector set includes a plurality of text vectors. A text vector is a unit of language that a computer can compute and thus use to perform clustering. Each sentence includes one or more text vectors. It is understood that two sentences that are similar may have a plurality of identical text vectors, but also necessarily have differences, for example, respectively include text vectors that are different from each other; alternatively, there are more text vectors for one sentence and fewer text vectors for another sentence.
Step S18, clustering the text vector set to form at least one interest class.
In this step, the text vector set is clustered by using, for example, a K-means clustering algorithm, and the clustering results can be integrated by using a clustering optimization (merging) algorithm known in the art, so as to obtain one or more interesting classes, where each interesting class is significantly different from other classes and contains at least one clause.
On the basis of the syntax tree pruning, the execution efficiency of the clustering algorithm and the accuracy of the clustering result are improved, so that the hot spots concerned by customers can be embodied more accurately.
According to a further improved embodiment of the present invention, after step S18, the following subsequent steps may also be performed. Specifically, for each class of interest, respectively: a1, extracting at least one hotspot keyword; a2, generating at least one hotspot expression based on the word frequency information of each hotspot keyword, wherein the hotspot expression corresponds to a permutation and combination of the at least one hotspot keyword; a3, determining the hot spot expression based on the support degree of each hot spot description in the document.
By way of example, an algorithm known to those skilled in the art (e.g., TextRank algorithm or TF-IDF algorithm) may be employed in a1 to extract hotspot keywords. In A2, selecting hot spot keywords with high word frequency, and forming different hot spot descriptions by utilizing various permutation and combination of the hot spot keywords; it will be appreciated that these hotspot descriptions are often not colloquially understandable or otherwise not grammatical. The combination of the keywords may not be arbitrary, but may conform to a certain abstract template, such as dvn (adverb-verb-noun) template. In A3, determining the support degree or the occurrence probability of various hotspot descriptions in the consulting document, and taking a plurality of hotspot descriptions with the highest support degree or occurrence probability as hotspot expressions finally displayed to the user; it will be appreciated that the hot spot expressions thus formed best fit the language habits of the client in the consulting document.
The second embodiment of the present invention provides a sentence hot-spot extraction system 20, which includes an emotion keyword extraction unit 201, a dependency syntax tree generation unit 202, a result syntax tree generation unit 211, a vector conversion unit 220, and a vector clustering unit 231.
The emotion keyword extraction unit 201 extracts at least one emotion keyword from the counsel document. Emotional keywords include, for example, "not," "none," "more," and
"lack".
The dependent syntax tree generating unit 202 performs a dependent syntax analysis on a first sentence selected from the consulting document to generate a dependent syntax tree. The first sentence may be any sentence in the consulting document or a sentence containing only the emotion keyword in the consulting document.
The result syntax tree generation unit 211 is coupled to the emotion keyword extraction unit 201 and the dependent syntax tree generation unit 202, respectively, and prunes the dependent syntax tree based on the first emotion keyword to form a result syntax tree. The first emotion keyword is derived from the first sentence, that is, the first emotion keyword corresponds to the first sentence when the first sentence includes the emotion keyword.
The vector conversion unit 220 forms a set of text vectors based on the result syntax tree.
The vector clustering unit 231 clusters the set of text vectors to form at least one class of interest, wherein each class of interest contains at least one clause and is significantly different from the other classes.
As a preferred embodiment, the statement hot-spot extraction system 20 further includes a hot-spot extraction unit (not shown in the drawings), and the hot-spot extraction unit respectively extracts at least one hot-spot keyword for each interested class; and generating at least one hotspot expression based on the word frequency information of each hotspot keyword, wherein the hotspot expression is obtained by arranging and combining different hotspot keywords. The hot spot extraction unit can further determine hot spot expressions based on the support degree of each hot spot description in the consulting document, and present the hot spot expressions to a user of the system.
The sentence hotspot extracting system 20 is not only suitable for hotspot extraction of a client consultation document, but also suitable for other sentence analysis occasions so as to efficiently and accurately acquire hotspots concerned by a client, and further, an intelligent response function can be started, for example, to respond. Alternatively, such a sentence hot spot extraction system may be integrated therein as a subsystem of an intelligent answering system or a gaming system or a voice recognition system, enabling rich functionality, facilitating a user experience.
In some embodiments of the invention, at least a portion of the system may be implemented using a distributed set of computing devices connected by a communications network, or may be implemented based on a "cloud". In such a system, multiple computing devices operate together to provide services by using their shared resources.
A "cloud" based implementation may provide one or more advantages, including: openness, flexibility and extensibility, centrally manageable, reliable, scalable, optimized for computing resources, having the ability to aggregate and analyze information across multiple users, connecting across multiple geographic areas, and the ability to use multiple mobile or data network operators for network connectivity.
The above description is only for the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Various modifications may be made by those skilled in the art without departing from the spirit of the invention and the appended claims.
Claims (10)
1. A sentence hot spot extraction method comprises the following steps:
a) extracting at least one emotion keyword from the document; wherein the document comprises at least one statement;
b) performing dependency syntax analysis on the first statement to generate a dependency syntax tree;
c) pruning the dependency syntax tree based on the first emotion keyword to form a result syntax tree; wherein the first emotion keyword is derived from the first sentence;
d) forming a text vector set based on the result syntax tree; and
e) and clustering the text vector set to form at least one interested class.
2. The method of claim 1, further comprising:
for each of the classes of interest:
extracting at least one hotspot keyword;
generating at least one hotspot description based on the word frequency information of each hotspot keyword, wherein the hotspot description corresponds to a permutation and combination of the at least one hotspot keyword; and
and determining hot spot expressions based on the support degree of each hot spot description in the document.
3. The method of claim 1, wherein step c) comprises:
c1) determining a root node of the dependency syntax tree;
c2) and performing breadth-first traversal based on the root node, and cutting out the nodes with the grammatical relation not conforming to the dependency white list and the subtrees of the nodes from the dependency syntax tree.
4. The method according to claim 3, characterized in that in step c1) the root node is determined by:
and searching upwards to the root node along the dependency syntax tree by taking the emotion keywords as the current searching leaf nodes, or determining the node of which the current parallel relation reaches the maximum depth limit value as the root node.
5. The method of claim 3, wherein the dependency white list comprises:
COO; a VOB; ADV; and, POB.
6. Method according to claim 1, characterized in that in step b):
and the grammatical relation and the part of speech of each node of the dependency syntax tree respectively meet the LTP specification and the ICTCCLAS standard.
7. The method according to claim 1, characterized in that step a) comprises in particular:
excluding from the document the sentences that do not contain any of the emotion keywords.
8. The method according to claim 1, wherein the clustering algorithm employed in step e) comprises: and (4) a K-means clustering algorithm.
9. The method of any one of claims 1-8, wherein the emotion keywords comprise:
"not"; "none"; "none".
10. A statement hotspot extraction system, comprising:
the emotion keyword extraction unit is used for extracting at least one emotion keyword from the document;
a dependency syntax tree generating unit for performing dependency syntax analysis on the first sentence to generate a dependency syntax tree; wherein the first sentence originates from the document;
a result syntax tree generating unit coupled to the emotion keyword extracting unit and the dependency syntax tree generating unit, respectively, and pruning the dependency syntax tree based on the first emotion keyword to form a result syntax tree; wherein the first emotion keyword is derived from the first sentence;
a vector conversion unit to form a set of text vectors based on the result syntax tree; and
and the vector clustering unit is used for clustering the text vector set to form at least one interested class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810871010.0A CN110852095B (en) | 2018-08-02 | 2018-08-02 | Statement hot spot extraction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810871010.0A CN110852095B (en) | 2018-08-02 | 2018-08-02 | Statement hot spot extraction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110852095A true CN110852095A (en) | 2020-02-28 |
| CN110852095B CN110852095B (en) | 2023-09-19 |
Family
ID=69595493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810871010.0A Active CN110852095B (en) | 2018-08-02 | 2018-08-02 | Statement hot spot extraction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110852095B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111414489A (en) * | 2020-03-25 | 2020-07-14 | 中金智汇科技有限责任公司 | Knowledge graph construction method and device, electronic equipment and readable storage medium |
| CN111460787A (en) * | 2020-03-27 | 2020-07-28 | 深圳价值在线信息科技股份有限公司 | Topic extraction method and device, terminal device and storage medium |
| CN111930953A (en) * | 2020-09-21 | 2020-11-13 | 北京合享智慧科技有限公司 | Text attribute feature identification, classification and structure analysis method and device |
| CN114036907A (en) * | 2021-11-18 | 2022-02-11 | 国网江苏省电力有限公司电力科学研究院 | Text data amplification method based on domain features |
| CN114662469A (en) * | 2022-02-25 | 2022-06-24 | 北京百度网讯科技有限公司 | Emotion analysis method and device, electronic equipment and storage medium |
| CN115017291A (en) * | 2022-08-04 | 2022-09-06 | 太平金融科技服务(上海)有限公司深圳分公司 | Hotspot problem analysis method and device, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110161071A1 (en) * | 2009-12-24 | 2011-06-30 | Metavana, Inc. | System and method for determining sentiment expressed in documents |
| CN102737013A (en) * | 2011-04-02 | 2012-10-17 | 三星电子（中国）研发中心 | Device and method for identifying statement emotion based on dependency relation |
| WO2015043075A1 (en) * | 2013-09-29 | 2015-04-02 | 广东工业大学 | Microblog-oriented emotional entity search system |
| CN106156041A (en) * | 2015-03-26 | 2016-11-23 | 科大讯飞股份有限公司 | Hot information finds method and system |
| CN106485525A (en) * | 2015-08-31 | 2017-03-08 | 百度在线网络技术（北京）有限公司 | Information processing method and device |
-
2018
- 2018-08-02 CN CN201810871010.0A patent/CN110852095B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110161071A1 (en) * | 2009-12-24 | 2011-06-30 | Metavana, Inc. | System and method for determining sentiment expressed in documents |
| CN102737013A (en) * | 2011-04-02 | 2012-10-17 | 三星电子（中国）研发中心 | Device and method for identifying statement emotion based on dependency relation |
| WO2015043075A1 (en) * | 2013-09-29 | 2015-04-02 | 广东工业大学 | Microblog-oriented emotional entity search system |
| CN106156041A (en) * | 2015-03-26 | 2016-11-23 | 科大讯飞股份有限公司 | Hot information finds method and system |
| CN106485525A (en) * | 2015-08-31 | 2017-03-08 | 百度在线网络技术（北京）有限公司 | Information processing method and device |
Non-Patent Citations (2)
| Title |
|---|
| 冯冲 等: "基于词汇语义和句法依存的情感关键句识别" * |
| 袁斌 等: "基于语义空间的藏文微博情感分析方法" * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111414489A (en) * | 2020-03-25 | 2020-07-14 | 中金智汇科技有限责任公司 | Knowledge graph construction method and device, electronic equipment and readable storage medium |
| CN111414489B (en) * | 2020-03-25 | 2023-10-27 | 中金智汇科技有限责任公司 | Knowledge graph construction method and device, electronic equipment and readable storage medium |
| CN111460787A (en) * | 2020-03-27 | 2020-07-28 | 深圳价值在线信息科技股份有限公司 | Topic extraction method and device, terminal device and storage medium |
| CN111460787B (en) * | 2020-03-27 | 2023-09-22 | 深圳价值在线信息科技股份有限公司 | Topic extraction method, topic extraction device, terminal equipment and storage medium |
| CN111930953A (en) * | 2020-09-21 | 2020-11-13 | 北京合享智慧科技有限公司 | Text attribute feature identification, classification and structure analysis method and device |
| CN111930953B (en) * | 2020-09-21 | 2021-02-02 | 北京合享智慧科技有限公司 | Text attribute feature identification, classification and structure analysis method and device |
| CN114036907A (en) * | 2021-11-18 | 2022-02-11 | 国网江苏省电力有限公司电力科学研究院 | Text data amplification method based on domain features |
| CN114662469A (en) * | 2022-02-25 | 2022-06-24 | 北京百度网讯科技有限公司 | Emotion analysis method and device, electronic equipment and storage medium |
| CN115017291A (en) * | 2022-08-04 | 2022-09-06 | 太平金融科技服务(上海)有限公司深圳分公司 | Hotspot problem analysis method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110852095B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110852095B (en) | Statement hot spot extraction method and system | |
| JP6309644B2 (en) | Method, system, and storage medium for realizing smart question answer | |
| CN113704451B (en) | Power user appeal screening method and system, electronic device and storage medium | |
| CN107402913B (en) | Method and device for determining antecedent | |
| US6243670B1 (en) | Method, apparatus, and computer readable medium for performing semantic analysis and generating a semantic structure having linked frames | |
| JP6998680B2 (en) | Interactive business support system and interactive business support program | |
| JP2001134575A (en) | Method and system for detecting frequently appearing pattern | |
| JP3266586B2 (en) | Data analysis system | |
| CN111831802B (en) | Urban domain knowledge detection system and method based on LDA topic model | |
| CN112035599B (en) | Query method and device based on vertical search, computer equipment and storage medium | |
| WO2017198031A1 (en) | Semantic parsing method and apparatus | |
| US20150178390A1 (en) | Natural language search engine using lexical functions and meaning-text criteria | |
| CN106960030A (en) | Pushed information method and device based on artificial intelligence | |
| KR20150096295A (en) | System and method for buinding q&as database, and search system and method using the same | |
| CA2823835A1 (en) | Voice search and response based on relevancy | |
| WO2014114175A1 (en) | Method and apparatus for providing search engine tags | |
| KR101654717B1 (en) | Method for producing structured query based on knowledge database and apparatus for the same | |
| CN106528768A (en) | Consultation hotspot analysis method and device | |
| CN110929007A (en) | Electric power marketing knowledge system platform and application method | |
| CN112347339A (en) | Search result processing method and device | |
| Putra et al. | A hybrid model for social media sentiment analysis for Indonesian text | |
| Olariu | Hierarchical clustering in improving microblog stream summarization | |
| Hong et al. | Comprehensive technology function product matrix for intelligent chatbot patent mining | |
| CN111046168A (en) | Method, apparatus, electronic device, and medium for generating patent summary information | |
| CN111325562A (en) | Grain safety tracing system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |