CN109087132B - Knowledge graph-based user question pushing method and device - Google Patents
Knowledge graph-based user question pushing method and device Download PDFInfo
- Publication number
- CN109087132B CN109087132B CN201810789092.4A CN201810789092A CN109087132B CN 109087132 B CN109087132 B CN 109087132B CN 201810789092 A CN201810789092 A CN 201810789092A CN 109087132 B CN109087132 B CN 109087132B
- Authority
- CN
- China
- Prior art keywords
- user
- keywords
- user behavior
- types
- degree
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000010276 construction Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a knowledge graph-based user problem pushing method and a knowledge graph-based user problem pushing device, wherein relevance analysis is performed on a plurality of problems with relevance, which are provided by a user, periodically, the relevance of user problems with different dimensions is considered, a preset number of high-correlation keywords corresponding to each type of user behavior keywords are finally obtained, the corresponding relation between each type of user behavior keywords and the high-correlation keywords is written into an established customer service center knowledge graph, and the relevance between the user problems and the user problems is accurately positioned. When a user question is received, a knowledge graph of the customer service center is called, potential problems of a plurality of users are automatically pushed for the user, the service quality of the customer service center is improved, and the user experience is improved.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to a user problem pushing method and device based on a knowledge graph.
Background
At present, many large enterprises, such as communication companies, power companies and banks, have special customer service centers so as to solve the problems brought by users.
The traditional customer service center only depends on the past experience of the manual seat and a limited knowledge base to answer the user questions, but the service capability of the manual seat is uneven, a certain time is needed for growth and promotion, and the updating condition of the knowledge base also depends on the execution condition of the regulation and regulation. Generally speaking, the service quality of the traditional customer service center depends on the professional degree of workers and the precision degree of the user description problem, so that the standard and precise service can not be provided for the user really, the user problem can not be positioned accurately, and the potential problem of the user can be pushed to the user in a personalized manner.
Disclosure of Invention
In view of the above, the invention provides a method and a device for pushing user problems based on a knowledge graph, so as to realize personalized pushing of potential problems of users.
In order to achieve the above purpose, the invention provides the following specific technical scheme:
a user question pushing method based on knowledge graph includes:
the method comprises the steps of obtaining problems to be analyzed, which are proposed by a plurality of users in a current period, wherein the problems to be analyzed comprise a plurality of problems with relevance;
analyzing each problem to be analyzed, and calculating the promotion degree of each two types of user behavior keywords in each problem to be analyzed, wherein the promotion degree represents the incidence relation between the user behavior keywords;
respectively counting the promotion degrees of every two types of user behavior keywords in a plurality of preset dimensions to obtain the promotion degrees and values of every two types of user behavior keywords in every preset dimension;
weighting and summing the promotion degree and the value of each two types of user behavior keywords in each preset dimension to obtain the total promotion degree of each two types of user behavior keywords;
for each type of user behavior keywords, obtaining a preset number of high-correlation keywords corresponding to each type of user behavior keywords according to the total promotion degree of each type of user behavior keywords corresponding to the user behavior keywords, and writing the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph;
when a user question is received, inquiring in the customer service center knowledge graph according to keywords in the user question to obtain a plurality of user potential questions comprising high-correlation keywords corresponding to the keywords in the user question;
pushing the plurality of user potential problems to a user.
Optionally, the method further includes:
and constructing a customer service center knowledge graph by using a big data technology according to the keywords in the preset field and the incidence relation among the keywords.
Optionally, the analyzing each problem to be analyzed and calculating the promotion of each two types of user behavior keywords in each problem to be analyzed includes:
analyzing each problem to be analyzed to obtain a plurality of user behavior keywords in the problem to be analyzed;
clustering a plurality of user behavior keywords in the problem to be analyzed to obtain a plurality of types of user behavior keywords;
respectively calculating the ratio of the number of times that each two types of user behavior keywords simultaneously appear in the same problem to the total number of problems in the problem to be analyzed to obtain the trust degree between each two types of user behavior keywords;
respectively calculating the ratio of the times of the two types of user behavior keywords appearing in the same problem at the same time and the times of any one type of user behavior keywords appearing in all the problems to be analyzed to obtain the support degree between the two types of user behavior keywords;
and respectively calculating the ratio of the trust degree to the support degree between every two types of user behavior keywords to obtain the promotion degree of every two types of user behavior keywords in the problem to be analyzed.
Optionally, the calculating the promotion degree of each two types of user behavior keywords in a plurality of preset dimensions respectively to obtain the promotion degree and the value of each two types of user behavior keywords in each preset dimension includes:
in the work order dimension, the promotion degree of every two types of user behavior keywords in the current period is counted to obtain the promotion degree and the value of every two types of user behavior keywords in the work order dimension;
in department dimension, counting the promotion degree of every two types of user behavior keywords in the current period to obtain the promotion degree and the value of every two types of user behavior keywords in the department dimension;
and in the personal dimension of the user, counting the promotion degrees of every two types of user behavior keywords in the current period to obtain the promotion degrees and values of every two types of user behavior keywords in the personal dimension of the user.
Optionally, the method further includes:
periodically counting the problems proposed by each user to obtain the high-frequency problem of each user;
when a user question is received, pushing a user's high frequency question of a plurality of user questions including a highly relevant keyword corresponding to a keyword of the user questions to the user.
Optionally, after the pushing the plurality of user questions with the highest degree of lift to the user, the method further includes:
and when the user clicks any one of the plurality of user potential questions, displaying an answer corresponding to the user question clicked by the user or displaying an answer link corresponding to the user question clicked by the user.
Optionally, after the pushing the plurality of user questions with the highest degree of lift to the user, the method further includes:
acquiring the click rate of each user question pushed by a user;
and updating the customer service center knowledge graph according to the click rate of each user question pushed by the user.
A knowledge-graph-based user question pushing device comprises:
the problem acquisition unit is used for acquiring problems to be analyzed, which are proposed by a plurality of users in the current period, and comprise a plurality of problems with relevance;
the lifting degree calculation unit is used for analyzing each problem to be analyzed and calculating the lifting degree of each two types of user behavior keywords in each problem to be analyzed, and the lifting degree represents the incidence relation between the user behavior keywords;
the system comprises a lifting degree counting unit, a lifting degree calculating unit and a display unit, wherein the lifting degree counting unit is used for counting the lifting degrees of every two types of user behavior key words in a plurality of preset dimensions respectively to obtain the lifting degrees and values of every two types of user behavior key words in each preset dimension;
the total promotion degree calculation unit is used for carrying out weighted summation on the promotion degree and the value of each two types of user behavior keywords in each preset dimension to obtain the total promotion degree of each two types of user behavior keywords;
the knowledge graph writing unit is used for obtaining a preset number of high-correlation keywords corresponding to each type of user behavior keywords according to the total promotion degree of each type of user behavior keywords corresponding to each type of user behavior keywords, and writing the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph;
the knowledge graph query unit is used for querying the knowledge graph of the customer service center according to the keywords in the user questions when receiving the user questions to obtain a plurality of user potential questions comprising the high-correlation keywords corresponding to the keywords in the user questions;
and the user problem pushing unit is used for pushing the plurality of user potential problems to the user.
Optionally, the apparatus further comprises:
and the knowledge graph construction unit is used for constructing the customer service center knowledge graph by using a big data technology according to the keywords in the preset field and the incidence relation among the keywords.
Optionally, the promotion calculation unit is specifically configured to, for each problem to be analyzed, analyze the problem to be analyzed to obtain a plurality of user behavior keywords in the problem to be analyzed; clustering a plurality of user behavior keywords in the problem to be analyzed to obtain a plurality of types of user behavior keywords; respectively calculating the ratio of the number of times that each two types of user behavior keywords simultaneously appear in the same problem to the total number of problems in the problem to be analyzed to obtain the trust degree between each two types of user behavior keywords; respectively calculating the ratio of the times of the two types of user behavior keywords appearing in the same problem at the same time and the times of any one type of user behavior keywords appearing in all the problems to be analyzed to obtain the support degree between the two types of user behavior keywords; and respectively calculating the ratio of the trust degree to the support degree between every two types of user behavior keywords to obtain the promotion degree of every two types of user behavior keywords in the problem to be analyzed.
Optionally, the promotion degree counting unit is specifically configured to count promotion degrees of every two types of user behavior keywords in the current period in the work order dimension to obtain promotion degrees and values of every two types of user behavior keywords in the work order dimension; in department dimension, counting the promotion degree of every two types of user behavior keywords in the current period to obtain the promotion degree and the value of every two types of user behavior keywords in the department dimension; and in the personal dimension of the user, counting the promotion degrees of every two types of user behavior keywords in the current period to obtain the promotion degrees and values of every two types of user behavior keywords in the personal dimension of the user.
Optionally, the apparatus further comprises:
the high-frequency problem counting unit is used for periodically counting the problems proposed by each user to obtain the high-frequency problem of each user;
the user question pushing unit is further used for pushing the user high-frequency questions in the plurality of user questions comprising the high-correlation keywords corresponding to the keywords in the user questions to the user when the user questions are received.
Optionally, the apparatus further comprises:
and the answer display unit is used for displaying the answer corresponding to the user question clicked by the user or displaying the answer link corresponding to the user question clicked by the user when the user clicks any one of the plurality of user potential questions.
Optionally, the apparatus further comprises:
the knowledge graph updating unit is used for acquiring the click rate of each user question pushed by the user; and updating the customer service center knowledge graph according to the click rate of each user question pushed by the user.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses a knowledge graph-based user problem pushing method and a knowledge graph-based user problem pushing device, wherein relevance analysis is carried out on a plurality of problems with relevance, which are proposed by a user, periodically, the relevance of user problems with different dimensions is considered, a preset number of high-correlation keywords corresponding to each type of user behavior keywords are finally obtained, the corresponding relation between each type of user behavior keywords and the high-correlation keywords is written into an established customer service center knowledge graph, and the relevance between the user problems and the user problems is accurately positioned. When a user question is received, a knowledge graph of the customer service center is called, potential problems of a plurality of users are automatically pushed for the user, the service quality of the customer service center is improved, and the user experience is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flow chart of a user problem pushing method based on a knowledge graph according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for calculating a lifting degree according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a user question pushing device based on a knowledge graph according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the embodiment discloses a user problem pushing method based on a knowledge graph, which is applied to a customer service center management system, where the customer service center management system may be a customer service center management system in any field, such as an intelligent power system ICT customer service center, a bank customer service center, and the like, and the method specifically includes the following steps:
s101: obtaining problems to be analyzed proposed by a plurality of users in the current period;
the duration of the period may be preset, such as one week, one month, etc., and the current period is the period of the current time.
The problem to be analyzed comprises a plurality of problems with relevance, the problem to be analyzed is a set of a plurality of problems proposed by a user in one business processing process, and the plurality of problems have relevance. For example, there is a correlation between the portal login problem and the collaborative office login problem.
The user questions can be presented by dialing a voice call to the customer service center or by the user interaction website. When the user question is in a speech mode, speech is converted to text using speech escape techniques.
S102: analyzing each problem to be analyzed, and calculating the promotion degree of each two types of user behavior keywords in each problem to be analyzed;
the promotion degree represents the incidence relation among the user behavior keywords.
Specifically, referring to fig. 2, for each problem to be analyzed, calculating the promotion degree of each two types of user behavior keywords specifically includes the following steps:
s201: analyzing the problem to be analyzed to obtain a plurality of user behavior keywords in the problem to be analyzed;
the problem to be analyzed can be analyzed by adopting a text mining technology to obtain a plurality of user behavior keywords.
S202: clustering a plurality of user behavior keywords in the problem to be analyzed to obtain a plurality of types of user behavior keywords;
the method for clustering the plurality of user behavior keywords in the problem to be analyzed can be any existing clustering method, and is not repeated herein.
It should be further noted that a class name is defined for each class of user behavior keywords obtained after clustering, a type can be defined for each class of user behavior keywords in a labeling manner, and a label is a class name.
S203: respectively calculating the ratio of the number of times that each two types of user behavior keywords simultaneously appear in the same problem to the total number of problems in the problem to be analyzed to obtain the trust degree between each two types of user behavior keywords;
s204: respectively calculating the ratio of the times of the two types of user behavior keywords appearing in the same problem at the same time and the times of any one type of user behavior keywords appearing in all the problems to be analyzed to obtain the support degree between the two types of user behavior keywords;
s205: and respectively calculating the ratio of the trust degree to the support degree between every two types of user behavior keywords to obtain the promotion degree of every two types of user behavior keywords in the problem to be analyzed.
S103: respectively counting the promotion degrees of every two types of user behavior keywords in a plurality of preset dimensions to obtain the promotion degrees and values of every two types of user behavior keywords in every preset dimension;
specifically, the preset dimensions are preset according to the service requirements of the customer service center, and taking the power system ICT customer service center as an example, the preset dimensions include: work order dimensions, department dimensions, and user personal dimensions.
In the work order dimension, the promotion degree of every two types of user behavior keywords in the current period is counted to obtain the promotion degree and the value of every two types of user behavior keywords in the work order dimension;
in department dimension, counting the promotion degree of every two types of user behavior keywords in the current period to obtain the promotion degree and the value of every two types of user behavior keywords in the department dimension;
and in the personal dimension of the user, counting the promotion degrees of every two types of user behavior keywords in the current period to obtain the promotion degrees and values of every two types of user behavior keywords in the personal dimension of the user.
S104: weighting and summing the promotion degree and the value of each two types of user behavior keywords in each preset dimension to obtain the total promotion degree of each two types of user behavior keywords;
it should be noted that, the weight of each preset dimension is preset, and the preset dimension includes: for example, the work order dimension, the department dimension, and the user individual dimension may be set to 0.2 for the weight of the work order dimension, 0.3 for the weight of the department dimension, and 0.5 for the weight of the user individual dimension.
S105: for each type of user behavior keywords, obtaining a preset number of high-correlation keywords corresponding to each type of user behavior keywords according to the total promotion degree of each type of user behavior keywords corresponding to the user behavior keywords, and writing the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph;
the preset number can be set according to business requirements, and if 5 potential problems of the users need to be pushed to the users, the preset number is set to be 5.
It should be noted that the method further includes:
and constructing a customer service center knowledge graph by using a big data technology according to the keywords in the preset field and the incidence relation among the keywords.
The most important function of the customer service center knowledge graph is an intelligent service engine, the intelligent service engine has dense algorithms and dense data, and the customer service center knowledge graph can be inquired according to the intelligent service engine.
S106: when a user question is received, inquiring in the customer service center knowledge graph according to keywords in the user question to obtain a plurality of user potential questions comprising high-correlation keywords corresponding to the keywords in the user question;
s107: pushing the plurality of user potential problems to a user.
The method for pushing the user potential problems to the user can be various, for example, when the user calls a call to a customer service center to ask a question, a worker queries in a knowledge graph of the customer service center according to the user problems to obtain a plurality of user potential problems, and the worker asks whether the user needs related problem service; when a user asks a question through the user interaction website, the website automatically displays a plurality of user potential questions according to the question of the user. For example, a high-frequency service such as "collaborative office" is searched, and a high-frequency service such as "how the collaborative office account authority is enabled" can be fed back, and a system knowledge integrated with the collaborative office system can also be fed back, for example, "how is a choice to select the filed information category when the digital archive system needs to file a file from the collaborative office? "and the like.
It should be noted that, when the user clicks any one of the plurality of user potential questions, the answer corresponding to the user potential question clicked by the user is displayed or the answer link corresponding to the user potential question clicked by the user is displayed.
The method for pushing user problems based on the knowledge graph disclosed by the embodiment periodically performs relevance analysis on a plurality of problems with relevance, which are provided by a user, considers the relevance of user problems with different dimensions, finally obtains a preset number of high-correlation keywords corresponding to each type of user behavior keywords, writes the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph, and accurately positions the relevance between the user problems and the user problems. When a user question is received, a knowledge graph of the customer service center is called, potential problems of a plurality of users are automatically pushed for the user, the service quality of the customer service center is improved, and the user experience is improved.
In order to enable the customer service center knowledge graph to more accurately locate the user problem and the association relationship between the user problems, the method for pushing the user problem based on the knowledge graph disclosed in this embodiment further includes:
periodically counting the problems proposed by each user to obtain the high-frequency problem of each user;
when a user question is received, pushing a user's high frequency question of a plurality of user questions including a highly relevant keyword corresponding to a keyword of the user questions to the user.
It can be understood that the high-frequency questions pushed to the user at this time are the high-frequency questions of the user, and the high-related keywords in the high-frequency questions correspond to the keywords in the user questions. The high-frequency problem of each user is obtained by periodically counting the problems proposed by each user, and the analysis of the user behavior track can be realized.
It should be further noted that, the high-frequency problem may also be pushed to the user first, and then the problems other than the high-frequency problem in all the potential problems of the user may be pushed to the user.
Another method for positioning the user problems and the association relationship between the user problems more accurately by the customer service center knowledge base comprises the following steps:
acquiring the click rate of each pushed user potential problem of the user;
and updating the customer service center knowledge graph according to the click rate of each user to each pushed potential problem.
Specifically, the association relationship between the user potential question with a high user click rate and the user behavior keyword corresponding to the user question is marked as high association in the customer service center knowledge graph, and the association relationship between the user potential question with a low user click rate and the user behavior keyword corresponding to the user question is marked as low association in the customer service center knowledge graph.
Through the mode, the self-learning updating of the customer service center knowledge map is realized, so that the customer service center knowledge map can more accurately position the user problems and the incidence relation among the user problems, the service quality of the customer service center is improved, and the user experience is improved.
Referring to fig. 3, the present embodiment discloses a method for pushing user questions based on a knowledge graph, which correspondingly discloses a device for pushing user questions based on a knowledge graph, including:
a question acquiring unit 301, configured to acquire a question to be analyzed, which is raised by multiple users in a current period and includes multiple questions with relevance;
a lifting degree calculation unit 302, configured to analyze each to-be-analyzed problem, and calculate a lifting degree of each two types of user behavior keywords in each to-be-analyzed problem, where the lifting degree represents an association relationship between the user behavior keywords;
optionally, the lifting degree calculating unit 302 is specifically configured to, for each problem to be analyzed, analyze the problem to be analyzed to obtain a plurality of user behavior keywords in the problem to be analyzed; clustering a plurality of user behavior keywords in the problem to be analyzed to obtain a plurality of types of user behavior keywords; respectively calculating the ratio of the number of times that each two types of user behavior keywords simultaneously appear in the same problem to the total number of problems in the problem to be analyzed to obtain the trust degree between each two types of user behavior keywords; respectively calculating the ratio of the times of the two types of user behavior keywords appearing in the same problem at the same time and the times of any one type of user behavior keywords appearing in all the problems to be analyzed to obtain the support degree between the two types of user behavior keywords; and respectively calculating the ratio of the trust degree to the support degree between every two types of user behavior keywords to obtain the promotion degree of every two types of user behavior keywords in the problem to be analyzed.
The promotion degree counting unit 303 is configured to count promotion degrees of each two types of user behavior keywords in a plurality of preset dimensions, respectively, to obtain promotion degrees and values of each two types of user behavior keywords in each preset dimension;
optionally, the promotion degree counting unit 303 is specifically configured to count promotion degrees of every two types of user behavior keywords in the current period in the work order dimension to obtain promotion degrees and values of every two types of user behavior keywords in the work order dimension; in department dimension, counting the promotion degree of every two types of user behavior keywords in the current period to obtain the promotion degree and the value of every two types of user behavior keywords in the department dimension; and in the personal dimension of the user, counting the promotion degrees of every two types of user behavior keywords in the current period to obtain the promotion degrees and values of every two types of user behavior keywords in the personal dimension of the user.
The total promotion degree calculation unit 304 is configured to perform weighted summation on the promotion degree and the value of each two types of user behavior keywords in each preset dimension to obtain a total promotion degree of each two types of user behavior keywords;
a knowledge graph writing unit 305, configured to, for each type of user behavior keyword, obtain a preset number of high-correlation keywords corresponding to each type of user behavior keyword according to a total promotion degree of each type of user behavior keyword corresponding thereto, and write a correspondence between each type of user behavior keyword and the high-correlation keywords into an established customer service center knowledge graph;
a knowledge graph query unit 306, configured to query the customer service center knowledge graph according to a keyword in a user question when a user question is received, so as to obtain a plurality of user potential questions including a highly relevant keyword corresponding to the keyword in the user question;
a user question pushing unit 307, configured to push the plurality of user potential questions to the user.
Optionally, the apparatus further comprises:
and the knowledge graph construction unit is used for constructing the customer service center knowledge graph by using a big data technology according to the keywords in the preset field and the incidence relation among the keywords.
Optionally, the apparatus further comprises:
the high-frequency problem counting unit is used for periodically counting the problems proposed by each user to obtain the high-frequency problem of each user;
the user question pushing unit is further used for pushing the user high-frequency questions in the plurality of user questions comprising the high-correlation keywords corresponding to the keywords in the user questions to the user when the user questions are received.
Optionally, the apparatus further comprises:
and the answer display unit is used for displaying the answer corresponding to the user potential question clicked by the user or displaying the answer link corresponding to the user potential question clicked by the user when the user clicks any one user question in the plurality of user potential questions.
Optionally, the apparatus further comprises:
the knowledge graph updating unit is used for acquiring the click rate of each pushed user potential problem of the user; and updating the customer service center knowledge graph according to the click rate of each user to each pushed potential problem.
The user question pushing device based on the knowledge graph disclosed by the embodiment periodically performs relevance analysis on a plurality of questions with relevance, which are provided by a user, considers the relevance of the user questions with different dimensions, finally obtains the preset number of high-correlation keywords corresponding to each type of user behavior keywords, writes the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph, and accurately positions the relevance between the user questions and the user questions. When a user question is received, a knowledge graph of the customer service center is called, potential problems of a plurality of users are automatically pushed for the user, the service quality of the customer service center is improved, and the user experience is improved.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A user question pushing method based on knowledge graph is characterized by comprising the following steps:
the method comprises the steps of obtaining problems to be analyzed, which are proposed by a plurality of users in a current period, wherein the problems to be analyzed comprise a plurality of problems with relevance;
analyzing each problem to be analyzed, and calculating the promotion degree of each two types of user behavior keywords in each problem to be analyzed, wherein the promotion degree represents the incidence relation between the user behavior keywords;
respectively counting the promotion degree of each two types of user behavior keywords in a plurality of preset dimensions to obtain the promotion degree and the value of each two types of user behavior keywords in each preset dimension, wherein the preset dimensions are preset according to the service requirements of the customer service center;
weighting and summing the promotion degree and the value of each two types of user behavior keywords in each preset dimension to obtain the total promotion degree of each two types of user behavior keywords;
for each type of user behavior keywords, obtaining a preset number of high-correlation keywords corresponding to each type of user behavior keywords according to the total promotion degree of each type of user behavior keywords corresponding to the user behavior keywords, and writing the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph;
when a user question is received, inquiring in the customer service center knowledge graph according to keywords in the user question to obtain a plurality of user potential questions comprising high-correlation keywords corresponding to the keywords in the user question;
pushing the plurality of user potential problems to a user;
the analyzing each problem to be analyzed and calculating the promotion degree of each two types of user behavior keywords in each problem to be analyzed comprise:
analyzing each problem to be analyzed to obtain a plurality of user behavior keywords in the problem to be analyzed;
clustering a plurality of user behavior keywords in the problem to be analyzed to obtain a plurality of types of user behavior keywords;
respectively calculating the ratio of the number of times that each two types of user behavior keywords simultaneously appear in the same problem to the total number of problems in the problem to be analyzed to obtain the trust degree between each two types of user behavior keywords;
respectively calculating the ratio of the times of the two types of user behavior keywords appearing in the same problem at the same time and the times of any one type of user behavior keywords appearing in all the problems to be analyzed to obtain the support degree between the two types of user behavior keywords;
and respectively calculating the ratio of the trust degree to the support degree between every two types of user behavior keywords to obtain the promotion degree of every two types of user behavior keywords in the problem to be analyzed.
2. The method of claim 1, further comprising:
and constructing a customer service center knowledge graph by using a big data technology according to the keywords in the preset field and the incidence relation among the keywords.
3. The method according to claim 1, wherein the step of counting the promotion degrees of each two types of user behavior keywords in a plurality of preset dimensions respectively to obtain the promotion degrees and values of each two types of user behavior keywords in each preset dimension comprises:
in the work order dimension, the promotion degree of every two types of user behavior keywords in the current period is counted to obtain the promotion degree and the value of every two types of user behavior keywords in the work order dimension;
in department dimension, counting the promotion degree of every two types of user behavior keywords in the current period to obtain the promotion degree and the value of every two types of user behavior keywords in the department dimension;
and in the personal dimension of the user, counting the promotion degrees of every two types of user behavior keywords in the current period to obtain the promotion degrees and values of every two types of user behavior keywords in the personal dimension of the user.
4. The method of claim 1, further comprising:
periodically counting the problems proposed by each user to obtain the high-frequency problem of each user;
when a user question is received, pushing a user's high frequency question of a plurality of user questions including a highly relevant keyword corresponding to a keyword of the user questions to the user.
5. The method of claim 1 or 4, wherein after said pushing the plurality of user questions to the user with the highest degree of lift, the method further comprises:
and when the user clicks any one of the plurality of user potential questions, displaying an answer corresponding to the user question clicked by the user or displaying an answer link corresponding to the user question clicked by the user.
6. The method of claim 1, wherein after said pushing the plurality of user questions to the user with the highest degree of lift, the method further comprises:
acquiring the click rate of each pushed user potential problem of the user;
and updating the customer service center knowledge graph according to the click rate of each user to each pushed potential problem.
7. A user question pushing device based on knowledge graph, comprising:
the problem acquisition unit is used for acquiring problems to be analyzed, which are proposed by a plurality of users in the current period, and comprise a plurality of problems with relevance;
the lifting degree calculation unit is used for analyzing each problem to be analyzed and calculating the lifting degree of each two types of user behavior keywords in each problem to be analyzed, and the lifting degree represents the incidence relation between the user behavior keywords;
the system comprises a promotion degree counting unit, a promotion degree calculating unit and a display unit, wherein the promotion degree counting unit is used for counting the promotion degrees of every two types of user behavior key words in a plurality of preset dimensions respectively to obtain the promotion degrees and values of every two types of user behavior key words in each preset dimension, and the preset dimensions are preset according to the service requirements of a customer service center;
the total promotion degree calculation unit is used for carrying out weighted summation on the promotion degree and the value of each two types of user behavior keywords in each preset dimension to obtain the total promotion degree of each two types of user behavior keywords;
the knowledge graph writing unit is used for obtaining a preset number of high-correlation keywords corresponding to each type of user behavior keywords according to the total promotion degree of each type of user behavior keywords corresponding to each type of user behavior keywords, and writing the corresponding relation between each type of user behavior keywords and the high-correlation keywords into the established customer service center knowledge graph;
the knowledge graph query unit is used for querying the knowledge graph of the customer service center according to the keywords in the user questions when receiving the user questions to obtain a plurality of user potential questions comprising the high-correlation keywords corresponding to the keywords in the user questions;
the user problem pushing unit is used for pushing the plurality of user potential problems to the user;
the promotion degree calculation unit is specifically used for analyzing each problem to be analyzed to obtain a plurality of user behavior keywords in the problems to be analyzed; clustering a plurality of user behavior keywords in the problem to be analyzed to obtain a plurality of types of user behavior keywords; respectively calculating the ratio of the number of times that each two types of user behavior keywords simultaneously appear in the same problem to the total number of problems in the problem to be analyzed to obtain the trust degree between each two types of user behavior keywords; respectively calculating the ratio of the times of the two types of user behavior keywords appearing in the same problem at the same time and the times of any one type of user behavior keywords appearing in all the problems to be analyzed to obtain the support degree between the two types of user behavior keywords; and respectively calculating the ratio of the trust degree to the support degree between every two types of user behavior keywords to obtain the promotion degree of every two types of user behavior keywords in the problem to be analyzed.
8. The device according to claim 7, wherein the promotion degree counting unit is specifically configured to count promotion degrees of every two types of user behavior keywords in a current period in a work order dimension to obtain promotion degrees and values of every two types of user behavior keywords in the work order dimension; in department dimension, counting the promotion degree of every two types of user behavior keywords in the current period to obtain the promotion degree and the value of every two types of user behavior keywords in the department dimension; and in the personal dimension of the user, counting the promotion degrees of every two types of user behavior keywords in the current period to obtain the promotion degrees and values of every two types of user behavior keywords in the personal dimension of the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810789092.4A CN109087132B (en) | 2018-07-18 | 2018-07-18 | Knowledge graph-based user question pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810789092.4A CN109087132B (en) | 2018-07-18 | 2018-07-18 | Knowledge graph-based user question pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109087132A CN109087132A (en) | 2018-12-25 |
| CN109087132B true CN109087132B (en) | 2021-07-30 |
Family
ID=64837736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810789092.4A Active CN109087132B (en) | 2018-07-18 | 2018-07-18 | Knowledge graph-based user question pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109087132B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110096B (en) * | 2019-05-13 | 2021-07-09 | 上海松鼠课堂人工智能科技有限公司 | Artificial intelligence K12 knowledge graph making management system and implementation method |
| CN110442700A (en) * | 2019-08-12 | 2019-11-12 | 珠海格力电器股份有限公司 | Man-machine more wheel dialogue methods and system, smart machine for human-computer interaction |
| CN110727786A (en) * | 2019-09-12 | 2020-01-24 | 武汉儒松科技有限公司 | Self-learning knowledge base management method and device, terminal device and storage medium |
| CN110689385A (en) * | 2019-10-16 | 2020-01-14 | 国网山东省电力公司信息通信公司 | Power customer service user portrait construction method based on knowledge graph |
| CN112667790B (en) * | 2020-12-18 | 2024-05-10 | 平安科技（深圳）有限公司 | Intelligent question-answering method, device, equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102685224B (en) * | 2012-04-28 | 2014-10-08 | 华为技术有限公司 | User behavior analysis method, related equipment and system |
| CN104462501A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Knowledge graph construction method and device based on structural data |
| US9514124B2 (en) * | 2015-02-05 | 2016-12-06 | International Business Machines Corporation | Extracting and recommending business processes from evidence in natural language systems |
| CN106919660A (en) * | 2017-02-09 | 2017-07-04 | 厦门快商通科技股份有限公司 | The clothes customer service intelligent Service method and system of knowledge based graphical spectrum technology |
| CN108021691B (en) * | 2017-12-18 | 2021-09-07 | 深圳前海微众银行股份有限公司 | Answer searching method, customer service robot and computer readable storage medium |
-
2018
- 2018-07-18 CN CN201810789092.4A patent/CN109087132B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109087132A (en) | 2018-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109087132B (en) | Knowledge graph-based user question pushing method and device | |
| US10318098B2 (en) | System and method of communication analysis | |
| US10789039B2 (en) | Call visualization | |
| Akhtar et al. | The potential of chatbots: analysis of chatbot conversations | |
| US20090307074A1 (en) | Crm system for enhanced retention of customers | |
| EP3899819B1 (en) | System and method of real-time wiki knowledge resources | |
| CN105512153A (en) | Method and device for service provision of online customer service system, and system | |
| CN110689385A (en) | Power customer service user portrait construction method based on knowledge graph | |
| Wang et al. | An integrated data analytics process to optimize data governance of non-profit organization | |
| CN105975479A (en) | Tag library-based telecom user interest degree analysis method and system | |
| CN107729549A (en) | A kind of robot client service method and system comprising elements recognition | |
| Asrani et al. | Diffusion and adoption of digital communications services in India | |
| Burkhard et al. | Extracting regular mobility patterns from sparse CDR data without a priori assumptions | |
| CN111400376A (en) | Method and device for building population analysis platform based on telecommunication data | |
| JP2007011541A (en) | Information server, information providing method, and program | |
| CN101179625A (en) | Voice number inquiry action based data analysis method and system | |
| US20150278768A1 (en) | Interviewing Aid | |
| US9210265B2 (en) | Exploring multiple contact channels to determine potential for reaching customers | |
| CN109167883B (en) | Intelligent shielding method and system based on social content and intelligent terminal | |
| CN102752462A (en) | Method and system for recommending telecommunications service | |
| CN109919811B (en) | Insurance agent culture scheme generation method based on big data and related equipment | |
| CN107862400A (en) | Scheduling method and device | |
| JP3976500B2 (en) | Reception support system | |
| CN112184419A (en) | Method and device for providing service for customer | |
| WO2017193528A1 (en) | Address book generation method and apparatus |
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 |