CN110555149A - Method, device and equipment for processing speech data and readable storage medium - Google Patents
Method, device and equipment for processing speech data and readable storage medium Download PDFInfo
- Publication number
- CN110555149A CN110555149A CN201910837223.6A CN201910837223A CN110555149A CN 110555149 A CN110555149 A CN 110555149A CN 201910837223 A CN201910837223 A CN 201910837223A CN 110555149 A CN110555149 A CN 110555149A
- Authority
- CN
- China
- Prior art keywords
- speech
- network
- expert
- wisdom
- data
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012545 processing Methods 0.000 title claims abstract description 31
- 238000004458 analytical method Methods 0.000 claims abstract description 148
- 238000012800 visualization Methods 0.000 claims abstract description 15
- 239000013598 vector Substances 0.000 claims description 28
- 238000003672 processing method Methods 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 17
- 238000013145 classification model Methods 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000003915 air pollution Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 239000010865 sewage Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
- 238000003911 water pollution Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Technology Law (AREA)
- Entrepreneurship & Innovation (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
the invention discloses a method, a device and equipment for processing speech data and a readable storage medium, wherein the method comprises the following steps: acquiring an utterance set about an ESG concern problem at each network platform; constructing an utterance network according to data relations among the utterance data in the utterance set; and analyzing the speech network to obtain a speech analysis result, and outputting the speech analysis result according to a preset visualization mode. The invention enables enterprises to intuitively obtain the speech opinions in the ESG aspect according to the speech analysis result visually displayed, thereby making and updating self-governing decisions according to the obtained speech opinions.
Description
Technical Field
The invention relates to the field of science and technology finance, in particular to a method, a device and equipment for processing speech data and a readable storage medium.
Background
The concept of responsible investment gradually gets wide acceptance of investors, and the responsible investment decides to select investment objects by analyzing the influence of enterprises on the environment, the fulfillment condition of Social responsibility and three major elements (ESG for short) of a company governing structure. In order to improve the appeal to investors, enterprises need to improve self-management and improve the performance of the three elements. At present, the ESG rating mechanism is mainly used for rating according to data actively disclosed by an enterprise, but a rating standard is not given, so that the enterprise cannot obtain specific guidance for the self administration of the enterprise through ESG rating. Although an enterprise can provide improved guidance opinions by inviting ESG experts, the way of engaging experts has many problems, such as expensive expense for engaging experts, large influence of human factors because the enterprise only needs a small amount of limited information for selecting the experts, no long-term parking of the experts, failure to update provided suggestions in time, deviation of opinions of one expert, and cost increase because of adoption of multi-expert opinions.
Disclosure of Invention
The invention mainly aims to provide a method, a device, equipment and a readable storage medium for processing talk data, and aims to solve the problem that at present, enterprises lack a better way for acquiring specific guidance in ESG (electronic service guide) to improve self governance.
In order to achieve the above object, the present invention provides a speech data processing method, including the following steps:
Acquiring an utterance set about an ESG concern problem at each network platform;
constructing an utterance network according to data relations among the utterance data in the utterance set;
And analyzing the speech network to obtain a speech analysis result, and outputting the speech analysis result according to a preset visualization mode.
Optionally, the step of acquiring an utterance set about an ESG concern at each network platform includes:
Acquiring a pre-collected wisdom library list and/or expert list of the ESG field;
Capturing the speech data of each wisdom and/or each expert on each network platform according to the wisdom list and/or the expert list;
And screening the speech data according to the preset concern problem to obtain a speech set about the ESG concern problem.
Optionally, the step of constructing an utterance network according to a data relationship between utterance data in the utterance set includes:
And constructing a language network by taking an intelligence library and/or an expert corresponding to each language data in the language set as nodes of the network and taking data relations among the language data in the language set as edges of the network.
optionally, the language analysis result includes a power analysis result and a leadership wisdom and/or a social account number of an expert, and the step of analyzing the language network to obtain the language analysis result includes:
Analyzing the speech network by adopting a preset authority analysis algorithm to obtain authority analysis results of each intellectual library and/or expert in the speech network;
determining a wisdom and/or an expert in leadership in the ESG concern problem according to the authority analysis result;
and acquiring the social account number of the wisdom and/or the expert in the leadership in each network platform.
Optionally, the speech analysis result includes a wisdom classification and/or an expert classification, and the step of analyzing the speech network to obtain the speech analysis result includes:
vectorizing each node in the speech network to obtain a feature vector of each node;
And classifying the feature vectors according to a preset classification model to obtain an intelligence library classification and/or an expert classification.
Optionally, the utterance analysis result further includes an opinion analysis result, and after the step of classifying each feature vector according to a preset classification model to obtain a wisdom classification and/or an expert classification, the method further includes:
and analyzing the wisdom-based language data and/or the expert-based language data under each wisdom-based category and/or expert category respectively according to a preset opinion analysis algorithm to obtain opinion analysis results, wherein the opinion analysis results comprise one or more items of opinion subject vocabularies, opinion relevance and opinion evolution processes.
Optionally, the language analysis result includes a community recognition result and/or a community evolution result, and the analyzing the language network to obtain the language analysis result includes:
carrying out community identification on the speech network according to a preset community identification algorithm to obtain a community identification result of an intelligence base and/or an expert; and/or the presence of a gas in the gas,
and carrying out community evolution analysis on the speech network according to a preset community evolution algorithm to obtain a community evolution result of the intelligence and/or experts.
In order to achieve the above object, the present invention also provides a speech data processing apparatus, including:
the acquisition module is used for acquiring an utterance set about the ESG concern problem on each network platform;
The network construction module is used for constructing the speech network according to the data relation among the speech data in the speech set;
and the analysis module is used for analyzing the speech network to obtain a speech analysis result and outputting the speech analysis result according to a preset visualization mode.
To achieve the above object, the present invention also provides a speech data processing apparatus including: a memory, a processor and a speech data processing program stored on the memory and executable on the processor, the speech data processing program, when executed by the processor, implementing the steps of the speech data processing method as described above.
Furthermore, in order to achieve the above object, the present invention also proposes a computer-readable storage medium having stored thereon a speech data processing program which, when executed by a processor, implements the steps of the speech data processing method as described above.
According to the invention, the language set about the ESG concern problem is acquired in each network platform, the language network is constructed according to the data relation of each language data in the language set, the language network is analyzed to obtain the language analysis result, and the language analysis result is output according to the preset visualization mode, so that an enterprise can intuitively acquire the language opinion in the ESG aspect according to the visually displayed language analysis result, and the self administration decision is made and updated according to the acquired language opinion.
drawings
FIG. 1 is a schematic diagram of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart diagram illustrating a first embodiment of a data processing method according to the present invention;
FIG. 3 is a schematic view of a speech data processing flow according to an embodiment of the present invention;
FIG. 4 is a schematic view of a speech data processing flow according to an embodiment of the present invention;
FIG. 5 is a block diagram of a data processing apparatus according to a preferred embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, fig. 1 is a schematic device structure diagram of a hardware operating environment according to an embodiment of the present invention.
it should be noted that, in the embodiment of the present invention, the data processing device may be a smart phone, a personal computer, a server, and the like, and is not limited herein.
As shown in fig. 1, the speech data processing apparatus may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the device architecture shown in fig. 1 does not constitute a limitation of the data processing device in question, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.
as shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a speech data processing program. Among them, the operating system is a program that manages and controls the hardware and software resources of the device, supporting the operation of the data processing program as well as other software or programs.
In the device shown in fig. 1, the user interface 1003 is mainly used for data communication with a client; the network interface 1004 is mainly used for establishing communication connection with each participating device; and the processor 1001 may be configured to call the speech data handler stored in the memory 1005 and perform the following operations:
acquiring an utterance set about an ESG concern problem at each network platform;
Constructing an utterance network according to data relations among the utterance data in the utterance set;
And analyzing the speech network to obtain a speech analysis result, and outputting the speech analysis result according to a preset visualization mode.
Further, the step of obtaining an argument set about the ESG concern at each network platform includes:
acquiring a pre-collected wisdom library list and/or expert list of the ESG field;
Capturing the speech data of each wisdom and/or each expert on each network platform according to the wisdom list and/or the expert list;
and screening the speech data according to the preset concern problem to obtain a speech set about the ESG concern problem.
Further, the step of constructing the speech network according to the data relationship among the speech data in the speech set includes:
And constructing a language network by taking an intelligence library and/or an expert corresponding to each language data in the language set as nodes of the network and taking data relations among the language data in the language set as edges of the network.
Further, the speech analysis result includes a power analysis result and a leadership wisdom and/or a social account number of an expert, and the step of analyzing the speech network to obtain the speech analysis result includes:
analyzing the speech network by adopting a preset authority analysis algorithm to obtain authority analysis results of each intellectual library and/or expert in the speech network;
determining a wisdom and/or an expert in leadership in the ESG concern problem according to the authority analysis result;
And acquiring the social account number of the wisdom and/or the expert in the leadership in each network platform.
further, the speech analysis result includes a wisdom classification and/or an expert classification, and the step of analyzing the speech network to obtain the speech analysis result includes:
Vectorizing each node in the speech network to obtain a feature vector of each node;
and classifying the feature vectors according to a preset classification model to obtain an intelligence library classification and/or an expert classification.
further, the utterance analysis result further includes an opinion analysis result, and after the step of classifying each feature vector according to a preset classification model to obtain a wisdom classification and/or an expert classification, the processor 1001 may be configured to call the utterance data processing program stored in the memory 1005, and further perform the following operations
and analyzing the wisdom-based language data and/or the expert-based language data under each wisdom-based category and/or expert category respectively according to a preset opinion analysis algorithm to obtain opinion analysis results, wherein the opinion analysis results comprise one or more items of opinion subject vocabularies, opinion relevance and opinion evolution processes.
further, the language analysis result includes a community recognition result and/or a community evolution result, and the step of analyzing the language network to obtain the language analysis result includes:
carrying out community identification on the speech network according to a preset community identification algorithm to obtain a community identification result of an intelligence base and/or an expert; and/or the presence of a gas in the gas,
And carrying out community evolution analysis on the speech network according to a preset community evolution algorithm to obtain a community evolution result of the intelligence and/or experts.
Based on the above structure, various embodiments of the speech data processing method are proposed.
Referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the data processing method according to the present invention.
while embodiments of the present invention provide embodiments of a data processing method, it should be noted that although a logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different than that shown or described herein. The execution subject of each embodiment of the data processing method of the present invention may be a device such as a smart phone, a personal computer, and a server, and for convenience of description, the system is used as the execution subject in the following embodiments. In this embodiment, the speech data processing method includes:
Step S10, obtaining an utterance set about the concern of the ESG on each network platform;
In this embodiment, in order to solve the problem that an enterprise lacks a specific guidance for acquiring an ESG to improve a self-administration way, a method for processing talk data is provided, talk data in the aspect of concern about the ESG by the enterprise is collected on each network platform, and big data analysis is performed on the talk data to obtain a visual display result, so that the enterprise can acquire intuitive talk opinions in the aspect of the ESG, and thus, the enterprise can be assisted in administering.
Specifically, the system can capture speech data about the concern about the ESG on each network platform in advance, and construct a speech set by using each speech data. Specifically, each network platform may be a microblog, a wechat public number, a notify, and the like. The ESG concern may be a concern about the ESG by an enterprise previously set in the system, such as environment, employee productivity, board structure, and the like. It should be understood that there are various manners for capturing the data about the ESG concern, and the capturing manner is not limited in this embodiment, for example, the capturing manner may be: the system takes preset ESG concern problems as key words, captures the speech issued by each social account number on each network platform, and takes the captured speech content and the data related to the speech as speech data for constructing a speech set. The set of utterances may be stored in the form of a database. The data related to the speech may include social account information for posting the speech, reference information of the speech, forwarding information of the speech, and the like. It should be understood that the language referred to in the present embodiment refers to a general term of opinion language, opinion article, etc., published through a network platform. It should be noted that the system may preset the update frequency of the language set, and capture the latest language data on each network platform according to the update frequency to update the language set.
Step S20, constructing a language network according to the data relationship among the language data in the language set;
the system constructs an utterance network according to the acquired utterance set about the ESG concern problem and the data relation among the utterance data in the utterance set. The system can determine the data relationship among the speech data according to the data related to the speech in the speech data, for example, determine the attention relationship among the social account numbers which issue the speech according to the social account number information in the speech data, and for example, determine the reference relationship among the speech according to the reference information of the speech in the speech data. The system can use each language as a network node to construct a language network according to the data relationship among the language data, and can also use a social account number which issues the language as a network node to construct the language network, for example, a knowledge graph can be constructed.
And step S30, analyzing the speech network to obtain a speech analysis result, and outputting the speech analysis result according to a preset visualization mode.
And after the system is constructed to obtain the speech network, analyzing the speech network to obtain a speech analysis result. It should be noted that there are various ways for the system to analyze the speech Network, and in this embodiment, the system may analyze the speech Network through a Network Representation Learning (Network Representation Learning) technology. Specifically, the system can represent the nodes in the speech network into a low-dimensional, real-valued and dense vector form through a network representation learning technology, the vector form can have the capability of representing and reasoning in a vector space, the system takes the vector form of each node as the input of a machine model, and the speech analysis result is output through the machine model. For example, when the machine model is a classification model and the speech network is a node of the speech network, the speech analysis result obtained by the system may be a classification result of each speech in the speech set, and the category may be a plurality of small concerns of the preset enterprise to the ESG, such as a sewage discharge problem in an environmental problem.
After the system obtains the speech analysis result, the speech analysis result can be output according to a preset visualization mode. The preset visualization mode may be a preset visualization mode, and the preset visualization mode may be different according to different speech analysis results, for example, when the speech analysis result is a classification result of speech, the visualization mode may be to display each category in a list form and support expansion display of each speech in the categories, and in addition, a query control may be further set in the system for a user to query each category or query the speech in each category.
In this embodiment, an utterance set related to the ESG concern is acquired in each network platform, an utterance network is constructed according to the data relationship of each utterance data in the utterance set, the utterance network is analyzed to obtain an utterance analysis result, and the utterance analysis result is output according to a preset visualization mode, so that an enterprise can intuitively acquire an utterance opinion in the ESG aspect according to the visualized utterance analysis result, and make and update a self administration decision according to the acquired utterance. Moreover, in the embodiment, the speech sets come from each network platform, so that the influence of human factors is reduced, the speech analysis result relates to the opinion of multi-party experts and is more objective and consistent, and the enterprise can make a perfect governing decision; after the system is successfully built, the personalized and automatic query of the enterprise can be realized, and compared with the method of engaging experts, the system only needs lower maintenance cost, so that the cost of the enterprise is reduced. The speech data in the system can be updated in time, so that wide expert opinions can be realized and can be quickly reflected on the governing decision of an enterprise.
Further, based on the first embodiment described above, a second embodiment of the data processing method according to the present invention is proposed, and in the second embodiment of the data processing method according to the present invention, the step S10 includes:
Step S101, acquiring a pre-collected wisdom library list and/or an expert list of the ESG field;
The wisdom base list or the expert list of the ESG field may be collected in advance, or the wisdom base list and the expert list may be collected. In the following embodiments, "a and/or B" is used to express "a and B" and the relationship of "a or B". The expert refers to an individual, and the intellectual property base mainly refers to a professional research institution which takes a public policy as a research object, influences government decisions as a research target, takes public interests as research guidance and takes social responsibility as a research criterion. The wisdom base list and/or the expert list in the ESG field can be acquired by means of providing by an organization for researching the wisdom base, pre-accumulating and sorting and the like. And uploading pre-collected wisdom base and/or expert list to the system. The system acquires a pre-collected wisdom base list and/or expert list of the ESG field.
step S102, capturing the speech data of each wisdom and/or each expert on each network platform according to the wisdom list and/or the expert list;
After the system acquires the wisdom base list and/or the expert list, the speech data of each wisdom base and/or each expert are captured on each network platform. Specifically, the system can capture the speech data of each wisdom library and/or expert in each network platform in a mode provided by the social platform combing, and can also capture the speech data of each wisdom library and/or each expert in each network platform in a mode of taking the full names and the Chinese and English acronyms of the wisdom libraries and/or the experts in the list as search keywords. The system can acquire all account numbers of each expert and/or each wisdom base and capture the speech data under all the account numbers to obtain all the speech data of each expert and/or each wisdom base on each network platform.
And S103, screening the speech data according to a preset concern problem to obtain a speech set about the ESG concern problem.
one or more ESG concern issues such as environmental issues, employee productivity issues, board structure issues, etc. may be preset in the system. The system screens the captured speech data of each wisdom library and/or expert according to the preset ESG problem to obtain a speech set about the ESG concern problem. Specifically, the system may filter, by using a text analysis technique in a Natural Language Processing (NLP) technique, the utterance data whose utterance contents are about the ESG concern, and use the filtered utterance data as an utterance set about the ESG concern. That is, the utterance data captured by the system according to the wisdom library and/or the expert list may or may not be utterance data related to the ESG concern problem, and therefore, all captured utterance data need to be screened, so that the utterance data in the finally obtained utterance set is utterance data related to the ESG concern problem, and the accuracy of the finally analyzed utterance analysis result is improved. Furthermore, when the preset concern questions are multiple, the system can classify the screened speech data according to the preset concern questions to obtain an environment-friendly influence corpus, an employee productivity corpus, a director structure corpus and the like.
In the embodiment, the most comprehensive ESG speech data is automatically acquired for enterprise users by acquiring the pre-acquired wisdom library list and/or expert list of the ESG field and capturing the speech data of each wisdom library and/or expert on each network platform according to the wisdom library list and/or the expert list, the speech data in the ESG aspect is automatically acquired for the enterprise users, the speech set related to the ESG concern problem is obtained by screening the speech data according to the preset concern problem, the most accurate ESG speech data is automatically acquired for the enterprise users, and therefore more accurate ESG speech analysis results are provided for the enterprise users, and the enterprise can obtain the most comprehensive and accurate enterprise governance guidance.
Further, based on the second embodiment, a third embodiment of the data processing method according to the present invention is proposed, and in the third embodiment of the data processing method according to the present invention, the step S20 includes:
Step S201, a wisdom and/or an expert corresponding to each language data in the language set are/is taken as nodes of a network, and a data relation among the language data in the language set is taken as an edge of the network to construct a language network.
After the system obtains the language set about the ESG concern problem, the intelligent library and/or the experts corresponding to the language data in the language set are/is taken as the nodes of the network, and the data relation among the language data in the language set is taken as the edge of the network to construct the language network. The intelligent library and/or the expert corresponding to each language data in the language set are/is an intelligent library or an expert which issues the language in the language data, and the intelligent library or the expert may correspond to a plurality of social account numbers and issue a plurality of languages in each social account number, so that a plurality of language data may correspond to the same intelligent library or the expert, and one intelligent library or one expert is taken as a node of the network, and the node corresponds to the language data which issues the language. The data relationships between the respective spoken data may include: and paying attention, quoting, forwarding and the like, taking the data relations as edges between nodes, such as an expert A node and an expert B node, wherein one language of the expert A refers to one language of the expert B, and an edge constructed on the basis of the quoting relation exists between the expert A node and the expert B node.
in this embodiment, the speech network is constructed by using the wisdom library and/or the experts corresponding to the speech data as the nodes of the network and using the data relationship between the speech data as the edges of the network, and the speech data is organized in the form of the network, so that the system can accurately and effectively analyze the speech data, and the speech analysis result in the ESG aspect favorable for enterprise governance is obtained.
Further, the speech analysis result includes a power analysis result and a leadership wisdom and/or a social account number of an expert, and the step S30 includes:
Step A10, analyzing the speech network by adopting a preset authority analysis algorithm to obtain authority analysis results of each wisdom and/or expert in the speech network;
after the speech network constructed by taking the wisdom and/or the experts as network nodes and taking the data relation between the speech data as the network edge is obtained, the system can analyze the speech network by adopting a preset power analysis algorithm to obtain the authority analysis result of each wisdom and/or expert in the speech network. The preset authority analysis algorithm may be a preset algorithm capable of analyzing Influence of nodes in the network, such as algorithm of PageRank, Influence mapping, and the like. The authority analysis result can be authority scores of all wisdom and/or experts as network nodes, and the higher the authority score is, the more authority the wisdom or the expert is.
Step A20, determining a wisdom and/or an expert in leading position in the ESG concern problem according to the authority analysis result;
The system determines a wisdom and/or an expert that is in leadership in the ESG concern based on the results of the authority analysis. Specifically, the leadership wisdom and/or expert refers to a powerful wisdom and/or expert. When the authority analysis result is the authority score of each wisdom and/or expert, the system can rank each wisdom and/or expert according to the authority score, rank the wisdom and/or expert with high authority in the front, and intercept a preset number of wisdom and/or expert from the ranking as the wisdom and/or expert in leadership. The preset number can be set as required, for example, 10.
And A30, acquiring the social account number of the wisdom and/or the expert in the leadership in each network platform.
And when the system acquires the wisdom and/or the expert in the leading position, acquiring the social account number of the wisdom and/or the expert in each network platform. The system can search the corresponding associated social account number according to the wisdom and/or the expert through the pre-stored association relationship between the wisdom and/or the expert and the social account number. After the social account numbers of the wisdom and/or the experts in the leadership are obtained, the system can output the names of the wisdom and/or the experts and the corresponding social account numbers in a visual mode.
In this embodiment, authority of each node in the speech network is analyzed by using an authority analysis algorithm to obtain an authority analysis result, a wisdom and/or an expert which is in a leading position in the ESG concern problem is determined according to the authority analysis result, a social account of the wisdom and/or the expert is obtained and output in a visual manner, so that an enterprise can intuitively obtain the wisdom and/or the expert which is in the leading position in the ESG concern problem, the wisdom and/or the expert can be concerned in each network platform according to the social account, and a governance decision of the enterprise is made and updated by paying attention to the published speech.
Further, the speech analysis result includes a wisdom classification and/or an expert classification, and step S30 further includes:
b10, vectorizing each node in the speech network to obtain the feature vector of each node;
after the speech network constructed by taking the wisdom and/or the experts as network nodes and taking the data relation between the speech data as the edge of the network is obtained, the system can carry out vectorization processing on each node in the speech network to obtain the characteristic vector of each node. The vectorization processing refers to extracting a plurality of characteristics of the nodes by adopting a vectorization processing algorithm, quantizing each characteristic, and forming a characteristic vector in a vector form by using the obtained quantization result of the multidimensional characteristics. Specifically, the system may preset a vectorization processing algorithm, such as a network embedding algorithm, to perform vectorization processing on each node in the speech network to obtain a feature vector of each wisdom and/or expert.
and B20, classifying the feature vectors according to a preset classification model to obtain a wisdom classification and/or an expert classification.
after the system obtains the feature vector of each node, classifying the feature vectors according to a preset classification model. The preset classification model may be a preset classification model, such as a K-nearest neighbor algorithm, a Support vector machine algorithm, and the like. The classification category may be preset, for example, a plurality of small concerns of the ESG concern problem are preset, and the respective wisdom libraries and/or experts are classified according to the small concerns through a classification model, that is, the small directions of research of each wisdom library and/or expert on the ESG concern problem are different, for example, a expert mainly issues some statements about the air pollution problem, B expert mainly issues the following statements about the water pollution problem, and a expert and B expert are classified into different categories. The classification category may also be set in advance, and the clustering algorithm, such as the K-nearest neighbor algorithm, is used to cluster the feature vectors, so as to classify the similar feature vectors into one category, and classify the dissimilar feature vectors into different categories, thereby finally obtaining the intellectual property classification and/or the expert classification.
Further, the speech analysis result further includes a opinion analysis result, and the step B20 further includes:
and step B30, analyzing the wisdom language data and/or the expert language data under each wisdom category and/or expert category respectively according to a preset opinion analysis algorithm to obtain opinion analysis results, wherein the opinion analysis results comprise one or more items in opinion theme vocabularies, opinion relevance and opinion evolution processes.
After the system obtains the wisdom base classification and/or the expert classification, each wisdom base class comprises at least one wisdom base, each expert class comprises at least one expert, and each wisdom base or expert corresponds to at least one speech data. The system can analyze the wisdom-based speech data and/or the expert-based speech data under each wisdom-based category and/or expert category respectively according to a preset opinion analysis algorithm to obtain opinion analysis results of each wisdom-based category and/or expert category, wherein the opinion analysis results can comprise one or more items of opinion subject words, opinion relevance and opinion evolution process. The preset viewpoint analysis algorithm may be a preset algorithm, such as a structural topic model, a probabilistic latent semantic analysis algorithm, and the like. The viewpoint topic vocabulary refers to topic vocabulary which is extracted from the opinion data and represents viewpoints, the viewpoint relevance refers to the relevance among the topic vocabularies of various viewpoints, and the evolution process of the viewpoints refers to the evolution process of the topic vocabularies of various viewpoints along with time.
After the system obtains the viewpoint analysis result, a wisdom classification and/or an expert classification can be output in a visual mode, for example, each category can be output in a list mode, the viewpoint analysis result of each category can be output based on the viewing operation of each category, for example, the viewpoint topic vocabulary can be output and displayed in a vocabulary cloud mode, the relevance and the evolution process of the viewpoint can be output and displayed in a network graph mode, nodes in the network graph can be the viewpoint topic vocabulary, and connecting lines among the nodes can represent the relevance and the evolution relation among the viewpoints.
In this embodiment, vectorization processing is performed on each node in the speech network to obtain feature vectors of each intelligence library and/or expert, each feature vector is classified according to a preset classification model to obtain intelligence library classification and/or expert classification, viewpoint analysis is performed on speech data in each category to obtain viewpoint analysis results, and the viewpoint analysis results are displayed, so that an enterprise can intuitively obtain small attention points, main viewpoints, relevance and evolution processes among the viewpoints, of each intelligence library and expert, in the ESG attention problem, so that the enterprise can quickly obtain expert viewpoints about the problem concerned by the enterprise, and management decisions of the enterprise are made and updated according to the expert viewpoints.
Further, based on the second or third embodiment, a fourth embodiment of the speech data processing method according to the present invention is proposed, in which the speech analysis result includes a community identification result and/or a community evolution result, and the step S30 includes:
Step C10, carrying out community recognition on the speech network according to a preset community recognition algorithm to obtain a community recognition result of the intelligence and/or experts;
after the language-theory network which is constructed by taking the intelligence library and/or the experts as network nodes and taking the data relation between the language-theory data as the network edge is obtained, the system can perform community identification on the language-theory network according to a preset community identification algorithm to obtain the community identification result of the intelligence library and/or the experts. An extreme example is illustrated for communities: if the speech network comprises 15 nodes, wherein 10 nodes are connected with each other with more edges, the other 5 nodes are connected with each other with more edges, and the front 10 nodes and the rear 5 nodes are not connected with each other, the front 10 nodes are identified as a community, and the rear 5 nodes are identified as a community. The preset community identification algorithm may be preset, for example, a Fast unfolding algorithm, a Labelpropagation algorithm, or the like is adopted. After the system obtains the community identification result, the system can visually output the community identification result, for example, the speech network is output in a network graph manner, and different colors are rendered for nodes belonging to different communities, so that enterprise users can distinguish the communities to which each intelligence and/or expert in the speech network belongs.
And step C20, carrying out community evolution analysis on the speech network according to a preset community evolution algorithm to obtain a community evolution result of the wisdom and/or the expert.
or the system can also perform community evolution analysis on the speaker network according to a preset community evolution algorithm to obtain a community evolution result of the intelligence and/or experts. The preset community evolution algorithm may be preset, for example, a predictive association model algorithm is adopted. Community evolution refers to the evolution process of a community formed by various wisdom bases and/or experts in a time sequence. After the system obtains the community evolution result, the community evolution result can be visually output, for example, an opinion network graph presenting different community divisions along with time change is output in a dynamic change graph mode, so that enterprise users can intuitively obtain the evolution process of each intelligence base and/or the focus of the expert in the ESG field according to the display result, and therefore a governance decision which is in line with the current situation is specified according to the evolution process.
It should be noted that the scheme for analyzing the speech network in steps a10, a20, and a30, the scheme for analyzing the speech network in steps B10, B20, and B30, and the scheme for analyzing the speech network in steps C10 and C20 in this embodiment may be implemented separately or in combination. The method has various combination modes, for example, the system can firstly carry out community identification and community evolution analysis on the speaker network; classifying the intelligence and/or experts in each community according to the community identification and community evolution results to obtain the classification of the intelligence and/or experts in each community; and then, performing viewpoint analysis on the speech data under each category, and performing authority analysis on the wisdom and/or experts under each category.
in an embodiment, the verbal analysis results may include any one or more of the results of the power analysis, social accounts of the leaders and/or wisdom and/or experts, wisdom and/or expert classifications, opinion analysis results, community identification results, and community evolution results. In one embodiment, the system may obtain and analyze the speech data according to the processing flow diagrams shown in fig. 3 and 4. Firstly, as shown in fig. 3, the system acquires a wisdom library and/or an expert list, and captures the speech published by the wisdom library and/or the expert on each network platform according to the list to obtain speech data; the system screens the speech data about the ESG concern problem from all speech data according to the preset concern problem of the enterprise to the ESG to form a speech set; then the system constructs a language network, such as a knowledge graph, for the language set; and then the system analyzes the language network, for example, analyzes the language network according to a network representation learning technology to obtain an authority analysis result, a community identification result, a classification result and a community evolution result of the intelligence library/expert. As shown in fig. 4, the system performs viewpoint analysis on the linguistic data of the wisdom and/or the experts of each category through a structure subject model algorithm according to the classification result and the community evolution result of the wisdom and/or the experts to obtain a viewpoint subject vocabulary, viewpoint relevance and viewpoint evolution result, and visually outputs the results for supporting the formulation and update of the administration decision of the enterprise in the aspect of the ESG.
in the embodiment, the community identification result and the community evolution result are obtained by carrying out the community identification and the community evolution analysis on the speech network, and the results are visually output, so that enterprise users can intuitively know the speech dynamics of each intelligence library and expert in the ESG field on the ESG concern problem, the enterprise users can more favorably and accurately find the speech and the viewpoint of the concern problem of the enterprise users from big data, and the governing decision beneficial to enterprise development is made or updated.
furthermore, an embodiment of the present invention further provides a speech data processing apparatus, and with reference to fig. 5, the speech data processing apparatus includes:
an obtaining module 10, configured to obtain, at each network platform, an utterance set related to an ESG concern;
A network construction module 20, configured to construct a speech network according to a data relationship between speech data in the speech set;
And the analysis module 30 is configured to analyze the speech network to obtain a speech analysis result, and output the speech analysis result according to a preset visualization manner.
further, the obtaining module 10 includes:
the system comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is used for acquiring a pre-acquired wisdom base list and/or an expert list of the ESG field;
The capturing unit is used for capturing the speech data of each intellectual library and/or each expert on each network platform according to the intellectual library list and/or the expert list;
And the screening unit is used for screening the speech data according to the preset concern problem to obtain a speech set about the ESG concern problem.
further, the network construction module 20 is configured to use the wisdom library and/or the experts corresponding to each piece of speech data in the speech set as nodes of the network, and use the data relationship between each piece of speech data in the speech set as an edge of the network to construct the speech network.
further, the speech analysis result includes a power analysis result and a leadership wisdom and/or a social account number of an expert, and the analysis module 30 includes:
the first analysis unit is used for analyzing the speech network by adopting a preset authority analysis algorithm to obtain authority analysis results of each wisdom and/or expert in the speech network;
the determining unit is used for determining a wisdom and/or an expert which is in a leading position in the ESG concern problem according to the authority analysis result;
and the acquisition unit is used for acquiring the social account numbers of the wisdom and/or the experts in the leadership in each network platform.
further, the speech analysis result includes a wisdom classification and/or an expert classification, and the analysis module 30 includes:
The vectorization processing unit is used for vectorizing each node in the speech network to obtain a feature vector of each node;
and the classification unit is used for classifying the feature vectors according to a preset classification model to obtain the intelligence library classification and/or the expert classification.
further, the speech analysis result further includes a viewpoint analysis result, and the analysis module 30 further includes:
And the second analysis unit is used for analyzing the wisdom-base speech data and/or the expert speech data under each wisdom-base category and/or expert category respectively according to a preset opinion analysis algorithm to obtain opinion analysis results, wherein the opinion analysis results comprise one or more items of opinion theme vocabularies, opinion relevance and opinion evolution process.
further, the utterance analysis result includes a community recognition result and/or a community evolution result, and the analysis module 30 includes:
the recognition unit is used for carrying out community recognition on the speech network according to a preset community recognition algorithm to obtain a community recognition result of the intelligence base and/or the expert; and/or the presence of a gas in the gas,
and the third analysis unit is used for carrying out community evolution analysis on the speech network according to a preset community evolution algorithm to obtain a community evolution result of the intelligence and/or experts.
the specific implementation of the speech data processing apparatus of the present invention is basically the same as the embodiments of the speech data processing method described above, and is not described herein again.
Furthermore, an embodiment of the present invention further provides a computer-readable storage medium, in which a speech data processing program is stored, and when being executed by a processor, the speech data processing program implements the steps of the speech data processing method described below.
the embodiments of the speech data processing device and the computer-readable storage medium of the present invention can refer to the embodiments of the speech data processing method of the present invention, and are not described herein again.
it should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
the above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A speech data processing method is characterized by comprising the following steps:
acquiring an utterance set about an ESG concern problem at each network platform;
constructing an utterance network according to data relations among the utterance data in the utterance set;
And analyzing the speech network to obtain a speech analysis result, and outputting the speech analysis result according to a preset visualization mode.
2. The method of claim 1, wherein the step of obtaining an argument set regarding the ESG concern at each network platform comprises:
acquiring a pre-collected wisdom library list and/or expert list of the ESG field;
Capturing the speech data of each wisdom and/or each expert on each network platform according to the wisdom list and/or the expert list;
And screening the speech data according to the preset concern problem to obtain a speech set about the ESG concern problem.
3. the method of claim 2, wherein the step of constructing a speech network based on data relationships between the speech data in the speech set comprises:
And constructing a language network by taking an intelligence library and/or an expert corresponding to each language data in the language set as nodes of the network and taking data relations among the language data in the language set as edges of the network.
4. the speech data processing method according to claim 3, wherein the speech analysis results comprise power analysis results and leadership wisdom and/or expert social account numbers, and the step of analyzing the speech network to obtain the speech analysis results comprises:
analyzing the speech network by adopting a preset authority analysis algorithm to obtain authority analysis results of each intellectual library and/or expert in the speech network;
Determining a wisdom and/or an expert in leadership in the ESG concern problem according to the authority analysis result;
and acquiring the social account number of the wisdom and/or the expert in the leadership in each network platform.
5. the speech data processing method according to claim 3, wherein the speech analysis result comprises a wisdom classification and/or an expert classification, and the step of analyzing the speech network to obtain the speech analysis result comprises:
vectorizing each node in the speech network to obtain a feature vector of each node;
And classifying the feature vectors according to a preset classification model to obtain an intelligence library classification and/or an expert classification.
6. The speech data processing method according to claim 5, wherein the speech analysis results further include opinion analysis results, and after the step of classifying each of the feature vectors according to a preset classification model to obtain the Chile classification and/or the expert classification, the method further comprises:
and analyzing the wisdom-based language data and/or the expert-based language data under each wisdom-based category and/or expert category respectively according to a preset opinion analysis algorithm to obtain opinion analysis results, wherein the opinion analysis results comprise one or more items of opinion subject vocabularies, opinion relevance and opinion evolution processes.
7. The method as claimed in claim 3, wherein the speech analysis results comprise community recognition results and/or community evolution results, and the step of analyzing the speech network to obtain the speech analysis results comprises:
carrying out community identification on the speech network according to a preset community identification algorithm to obtain a community identification result of an intelligence base and/or an expert; and/or the presence of a gas in the gas,
And carrying out community evolution analysis on the speech network according to a preset community evolution algorithm to obtain a community evolution result of the intelligence and/or experts.
8. a speech data processing apparatus, characterized in that the speech data processing apparatus comprises:
The acquisition module is used for acquiring an utterance set about the ESG concern problem on each network platform;
The network construction module is used for constructing the speech network according to the data relation among the speech data in the speech set;
And the analysis module is used for analyzing the speech network to obtain a speech analysis result and outputting the speech analysis result according to a preset visualization mode.
9. a speech data processing apparatus, characterized in that the speech data processing apparatus comprises: memory, a processor and a speech data processing program stored on the memory and executable on the processor, which speech data processing program, when executed by the processor, carries out the steps of the speech data processing method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a speech data processing program is stored on the computer-readable storage medium, which when executed by a processor implements the steps of the speech data processing method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910837223.6A CN110555149A (en) | 2019-09-05 | 2019-09-05 | Method, device and equipment for processing speech data and readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910837223.6A CN110555149A (en) | 2019-09-05 | 2019-09-05 | Method, device and equipment for processing speech data and readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110555149A true CN110555149A (en) | 2019-12-10 |
Family
ID=68739166
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910837223.6A Pending CN110555149A (en) | 2019-09-05 | 2019-09-05 | Method, device and equipment for processing speech data and readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110555149A (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103024017A (en) * | 2012-12-04 | 2013-04-03 | 武汉大学 | Method for distinguishing important goals and community groups of social network |
CN103793489A (en) * | 2014-01-16 | 2014-05-14 | 西北工业大学 | Method for discovering topics of communities in on-line social network |
JP2014219985A (en) * | 2013-05-09 | 2014-11-20 | 富士通株式会社 | Field opinion leader extraction method, and field opinion leader extraction device |
CN104199947A (en) * | 2014-09-11 | 2014-12-10 | 浪潮集团有限公司 | Important person speech supervision and incidence relation excavating method |
CN105354343A (en) * | 2015-12-24 | 2016-02-24 | 成都陌云科技有限公司 | User characteristic mining method based on remote dialogue |
CN105608194A (en) * | 2015-12-24 | 2016-05-25 | 成都陌云科技有限公司 | Method for analyzing main characteristics in social media |
CN107239498A (en) * | 2017-05-03 | 2017-10-10 | 同济大学 | A kind of method for excavating overlapping community's dynamic evolution correlation rule |
US20180067910A1 (en) * | 2016-09-06 | 2018-03-08 | Microsoft Technology Licensing, Llc | Compiling Documents Into A Timeline Per Event |
CN107835113A (en) * | 2017-07-05 | 2018-03-23 | 中山大学 | Abnormal user detection method in a kind of social networks based on network mapping |
CN108197224A (en) * | 2017-12-28 | 2018-06-22 | 广州虎牙信息科技有限公司 | User group sorting technique, storage medium and terminal |
CN109951377A (en) * | 2019-03-20 | 2019-06-28 | 西安电子科技大学 | A kind of good friend's group technology, device, computer equipment and storage medium |
-
2019
- 2019-09-05 CN CN201910837223.6A patent/CN110555149A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103024017A (en) * | 2012-12-04 | 2013-04-03 | 武汉大学 | Method for distinguishing important goals and community groups of social network |
JP2014219985A (en) * | 2013-05-09 | 2014-11-20 | 富士通株式会社 | Field opinion leader extraction method, and field opinion leader extraction device |
CN103793489A (en) * | 2014-01-16 | 2014-05-14 | 西北工业大学 | Method for discovering topics of communities in on-line social network |
CN104199947A (en) * | 2014-09-11 | 2014-12-10 | 浪潮集团有限公司 | Important person speech supervision and incidence relation excavating method |
CN105354343A (en) * | 2015-12-24 | 2016-02-24 | 成都陌云科技有限公司 | User characteristic mining method based on remote dialogue |
CN105608194A (en) * | 2015-12-24 | 2016-05-25 | 成都陌云科技有限公司 | Method for analyzing main characteristics in social media |
US20180067910A1 (en) * | 2016-09-06 | 2018-03-08 | Microsoft Technology Licensing, Llc | Compiling Documents Into A Timeline Per Event |
CN107239498A (en) * | 2017-05-03 | 2017-10-10 | 同济大学 | A kind of method for excavating overlapping community's dynamic evolution correlation rule |
CN107835113A (en) * | 2017-07-05 | 2018-03-23 | 中山大学 | Abnormal user detection method in a kind of social networks based on network mapping |
CN108197224A (en) * | 2017-12-28 | 2018-06-22 | 广州虎牙信息科技有限公司 | User group sorting technique, storage medium and terminal |
CN109951377A (en) * | 2019-03-20 | 2019-06-28 | 西安电子科技大学 | A kind of good friend's group technology, device, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20210232762A1 (en) | Architectures for natural language processing | |
CN111444428B (en) | Information recommendation method and device based on artificial intelligence, electronic equipment and storage medium | |
CN111753060B (en) | Information retrieval method, apparatus, device and computer readable storage medium | |
KR102288249B1 (en) | Information processing method, terminal, and computer storage medium | |
CN110888990B (en) | Text recommendation method, device, equipment and medium | |
CN112632385A (en) | Course recommendation method and device, computer equipment and medium | |
CN112800170A (en) | Question matching method and device and question reply method and device | |
CN108885623A (en) | The lexical analysis system and method for knowledge based map | |
CN109299245B (en) | Method and device for recalling knowledge points | |
CN111098312A (en) | Window government affairs service robot | |
US20140052445A1 (en) | Voice search and response based on relevancy | |
CN109508441B (en) | Method and device for realizing data statistical analysis through natural language and electronic equipment | |
US11966698B2 (en) | System and method for automatically tagging customer messages using artificial intelligence models | |
CN114077705A (en) | Method and system for portraying media account on social platform | |
Wei et al. | Online education recommendation model based on user behavior data analysis | |
CN112163415A (en) | User intention identification method and device for feedback content and electronic equipment | |
CN114445043B (en) | Open ecological cloud ERP-based heterogeneous graph user demand accurate discovery method and system | |
CN115660695A (en) | Customer service personnel label portrait construction method and device, electronic equipment and storage medium | |
CN110555149A (en) | Method, device and equipment for processing speech data and readable storage medium | |
CN114090777A (en) | Text data processing method and device | |
CN114153948A (en) | Question-answer knowledge base construction method, intelligent interaction method and device | |
US20230108518A1 (en) | Information processing apparatus, information processing method, and program | |
CN116431779B (en) | FAQ question-answering matching method and device in legal field, storage medium and electronic device | |
US11907500B2 (en) | Automated processing and dynamic filtering of content for display | |
CN114036373B (en) | Searching method and device, electronic equipment and storage medium |
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 |