CN109885651B - Question pushing method and device - Google Patents
Question pushing method and device Download PDFInfo
- Publication number
- CN109885651B CN109885651B CN201910038458.9A CN201910038458A CN109885651B CN 109885651 B CN109885651 B CN 109885651B CN 201910038458 A CN201910038458 A CN 201910038458A CN 109885651 B CN109885651 B CN 109885651B
- Authority
- CN
- China
- Prior art keywords
- standard
- queried
- similarity
- target
- question
- 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 28
- 238000013507 mapping Methods 0.000 claims abstract description 56
- 239000013598 vector Substances 0.000 claims description 87
- 230000006399 behavior Effects 0.000 claims description 78
- 238000012216 screening Methods 0.000 claims description 30
- 238000004364 calculation method Methods 0.000 claims description 24
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a problem pushing method and device. The invention relates to the field of artificial intelligence, which comprises the following steps: acquiring behavior data of a target user; searching a standard problem corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problem stored in the target database; the first criteria question is pushed to the target user. Therefore, the technical scheme provided by the embodiment of the invention can solve the problem that the intelligent customer service system in the prior art cannot intelligently recommend the problem to the user.
Description
[ Field of technology ]
The invention relates to the field of artificial intelligence, in particular to a problem pushing method and device.
[ Background Art ]
The intelligent customer service system is a technology developed on the basis of large-scale knowledge processing, and provides a quick and effective technical means based on natural language for communication between enterprises and massive users.
In order for the intelligent customer service system to conveniently provide services for users, a set of standard questions and answers corresponding to the standard questions need to be stored in the intelligent customer service system, when the users ask questions, the standard questions input by the users and the answers corresponding to the standard questions are found, for example, "how to change mobile phone short message service? "this standard question, the intelligent customer service system pushes an answer to this standard question to the user.
However, in practical application, for example, a user encounters difficulty in using a certain software to ask questions, but the user may not accurately describe the questions, the questions input by the user are not completely consistent with the standard questions, the intelligent customer service system extracts keywords from the questions input by the user, searches for more than one standard question according to the keyword search standard questions, the intelligent customer service system returns all the searched standard questions to the user, the user needs to consume more time to find and confirm the questions really wanting to ask questions from a plurality of standard questions returned by the intelligent customer service system, and then the intelligent customer service system returns answers to the questions.
[ Invention ]
In view of the above, the embodiment of the invention provides a problem pushing method and device, which are used for solving the problem that the intelligent customer service system in the prior art cannot intelligently recommend the problem to a user.
In one aspect, an embodiment of the present invention provides a problem pushing method, where the method includes: acquiring behavior data of a target user; searching a standard problem corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problem stored in the target database, and the preset mapping relation is determined in advance through the following steps: receiving a problem inquiry request sent by a plurality of users, and collecting behavior data generated when the plurality of users send the problem inquiry request, wherein the problem inquiry request comprises a problem to be inquired; extracting the problems to be queried contained in the problem query request, and screening out the standard problems with highest similarity with the problems to be queried; establishing a mapping relation between the collected behavior data and the screened standard problems; pushing the first standard question to the target user.
Further, the screening the standard problem with the highest similarity with the to-be-queried problem includes: calculating the distance between the sentence vector corresponding to the problem to be queried and the cluster center point of each cluster category in N cluster categories to obtain N distances, wherein the N cluster categories are determined in advance according to the following steps: acquiring all standard problems in the target database; respectively converting all the standard questions into corresponding sentence vectors; clustering sentence vectors corresponding to all the standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2; screening out the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category; respectively calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category to obtain M similarity, wherein M is the number of standard problems in the target cluster category; and taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the to-be-queried problem.
Further, the calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category includes: determining a first target word and a second target word, wherein the first target word is a word which appears in the to-be-queried problem and does not appear in a second standard problem, the second target word is a word which appears in the second standard problem and does not appear in the to-be-queried problem, and the second standard problem is any standard problem included in the target cluster category; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem, the first target word and the second target word.
Further, the calculating the similarity between the query question and the second standard question according to the sentence vectors respectively corresponding to the query question and the second standard question and the first target term and the second target term includes: calculating a similarity degree parameter between the first target word and the second target word; and calculating the similarity between the to-be-queried problem and the second standard problem according to sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem and the similarity degree parameter.
Further, the calculating the similarity between the question to be queried and the second standard question according to the sentence vectors and the similarity parameters respectively corresponding to the question to be queried and the second standard question includes: according to the formulaCalculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Representing sentence vector corresponding to the second standard question,/>Modulus of sentence vector corresponding to the question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
In one aspect, an embodiment of the present invention provides a problem pushing apparatus, including: the acquisition unit is used for acquiring behavior data of the target user; the searching unit is used for searching standard questions corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard question, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard questions stored in the target database; a pushing unit, configured to push the first standard problem to the target user, where the preset mapping relationship is determined in advance by a determining unit, and the determining unit includes: a receiving subunit, configured to receive question query requests sent by a plurality of users; the acquisition subunit is used for acquiring behavior data generated when the plurality of users send the problem inquiry requests, wherein the problem inquiry requests comprise problems to be inquired; an extracting subunit, configured to extract a problem to be queried included in the problem query request; the screening subunit is used for screening out the standard problem with highest similarity with the problem to be queried; and the building subunit is used for building a mapping relation between the collected behavior data and the screened standard problems.
Further, the screening subunit includes: the first calculation module is used for calculating the distance between the sentence vector corresponding to the to-be-queried problem and the cluster center point of each cluster type in the N cluster types to obtain N distances, wherein the N cluster types are determined in advance according to the following steps: acquiring all standard problems in the target database; respectively converting all the standard questions into corresponding sentence vectors; clustering sentence vectors corresponding to all the standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2; the screening module is used for screening the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category; the second calculation module is used for calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category respectively to obtain M similarity, wherein M is the number of standard problems in the target cluster category; and the determining module is used for taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried.
Further, the second computing module includes: a determining submodule, configured to determine a first target term and a second target term, where the first target term is a term that appears in the question to be queried and does not appear in a second standard question, the second target term is a term that appears in the second standard question and does not appear in the question to be queried, and the second standard question is any standard question included in the target cluster category; and the calculation sub-module is used for calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem, the first target word and the second target word.
Further, the computing submodule includes: the first calculation large module is used for calculating similarity parameters between the first target word and the second target word; and the second calculation big module is used for calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem and the similarity degree parameter.
Further, the second calculation large module includes: a calculation small module for calculating the formulaCalculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Representing sentence vector corresponding to the second standard question,/>Modulus of sentence vector corresponding to the question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
In one aspect, an embodiment of the present invention provides a storage medium, where the storage medium includes a stored program, where when the program runs, the device where the storage medium is controlled to execute the problem pushing method described above.
In one aspect, an embodiment of the present invention provides a computer device, including a memory for storing information including program instructions, and a processor for controlling execution of the program instructions, where the program instructions, when loaded and executed by the processor, implement the steps of the problem pushing method described above.
In the embodiment of the invention, the mapping relation between the behavior data of the user and the standard problem is determined in advance according to the behavior data of the historical user, and the behavior data of the target user is obtained; according to the mapping relation between the behavior data of the user and the standard questions stored in the target database, searching the standard questions corresponding to the behavior data of the target user to obtain a first standard question; the first standard questions are pushed to the target users, and the problems that the target users possibly ask questions are predicted according to the behavior data of the target users, so that the effect of intelligently recommending the problems to the users is achieved, and the problem that the intelligent customer service system in the prior art cannot intelligently recommend the problems to the users is solved.
[ Description of the drawings ]
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an alternative problem pushing method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an alternative problem pushing device according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an alternative computer device provided by an embodiment of the present invention.
[ Detailed description ] of the invention
For a better understanding of the technical solution of the present invention, the following detailed description of the embodiments of the present invention refers to the accompanying drawings.
It should be understood that the described embodiments are merely some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be understood that the term "and/or" as used herein is merely one relationship describing the association of the associated objects, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.
FIG. 1 is a flowchart of an alternative problem pushing method according to an embodiment of the present invention, as shown in FIG. 1, the method includes:
step S102, behavior data of a target user is obtained.
The behavior data of the target user includes at least one of operation data, state data and help data generated by the target user during the process of using the target software, for example, the following data can be used as the behavior data of the target user: the target user triggers the behavior data of a control in the target software, the target user triggers the behavior data of a link address in the target software, the target user uses the time information state data of the target software, and the target user sends help data of customer service consultation in the process of using the target software, and the like.
Step S104, searching a standard problem corresponding to the behavior data of the target user according to a preset mapping relation, and obtaining a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problem stored in the target database, and the preset mapping relation is determined in advance through the following steps: receiving a problem inquiry request sent by a plurality of users, and collecting behavior data generated when the plurality of users send the problem inquiry request, wherein the problem inquiry request comprises a problem to be inquired; extracting the problems to be queried contained in the problem query request, and screening out the standard problems with highest similarity with the problems to be queried; and establishing a mapping relation between the collected behavior data and the screened standard problems.
A set of standard questions is stored in a database of the intelligent customer service system, and the embodiment of the invention establishes a mapping relation between the behavior data of the user and the standard questions stored in the database based on the behavior data of the historical user and the question query request sent by the historical user, can match the corresponding standard questions according to the behavior data of the target user, and pushes the standard questions for the target user. After the target user sees the standard question, the standard question can be confirmed to confirm that the question is really the question which the target user wants to search, and after the target user confirms, the target user is pushed the answer of the standard question.
The mapping relationship between the behavior data of the user and the standard questions stored in the target database is pre-established, and in the following, it is specifically described how to establish the mapping relationship between the behavior data of the user and the standard questions stored in the target database.
For example: the history user A encounters a difficult problem when using the target software, at this time, the history user A can send a problem inquiry request to customer service of the target software, the problem inquiry request can be in a telephone form or can be an online consultation, and when receiving the problem inquiry request sent by the history user A, the intelligent customer service system determines behavior data D1 of the history user A when using the target software. At this time, the intelligent customer service system screens out the standard problem Q1 with the highest similarity to the problem to be queried among the standard problems stored in the database according to the problem to be queried included in the problem query request sent by the historical user a, so that there may be a mapping relationship between the behavior data D1 and the standard problem Q1, but at this time, it cannot be determined that there is a mapping relationship between the behavior data D1 and the standard problem Q1, because the behaviors of the individual users do not have universality.
According to the method, a large number of historical user question query requests and behavior data at the same time or before the question query requests are collected, standard questions with highest similarity with the questions to be queried are screened out of standard questions stored in a database according to questions to be queried contained in the question query requests sent by the large number of historical users, and if a certain behavior data corresponds to a certain standard question under most conditions, a mapping relation exists between the behavior data and the standard questions. For example, a total of 10000 historical users' question query requests and behavior data at the same time or before the question query requests are collected, and it is found that standard questions corresponding to questions to be queried contained in the question query requests sent by 9000 historical users are all standard questions Q1 (standard questions Q1 are standard questions with highest similarity to questions to be queried among all standard questions stored in a database), and the behavior data of 9000 historical users at the same time or before the question query requests are all behavior data D1, which indicates that a mapping relationship exists between the behavior data D1 and the standard questions Q1.
Step S106, pushing the first standard questions to the target user.
In the embodiment of the invention, the mapping relation between the behavior data of the user and the standard problem is determined in advance according to the behavior data of the historical user, and the behavior data of the target user is obtained; according to the mapping relation between the behavior data of the user and the standard questions stored in the target database, searching the standard questions corresponding to the behavior data of the target user to obtain a first standard question; the first standard questions are pushed to the target users, and the problems that the target users possibly ask questions are predicted according to the behavior data of the target users, so that the effect of intelligently recommending the problems to the users is achieved, and the problem that the intelligent customer service system in the prior art cannot intelligently recommend the problems to the users is solved.
Optionally, screening the standard problem with the highest similarity with the problem to be queried includes: calculating distances between sentence vectors corresponding to the to-be-queried problem and cluster center points of each of N cluster categories to obtain N distances, wherein the N cluster categories are determined in advance according to the following steps: acquiring all standard problems in a target database; all standard questions are respectively converted into corresponding sentence vectors; clustering sentence vectors corresponding to all standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2; screening out the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category; respectively calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category to obtain M similarity, wherein M is the number of standard problems in the target cluster category; and taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried.
As an optional implementation manner, the step of clustering sentence vectors corresponding to all standard questions to obtain N cluster categories includes: s1, determining an N value according to priori experience, wherein N is the cluster number of the clusters; s2, randomly selecting sentence vectors corresponding to N standard problems as cluster center points of N cluster categories; s3, for the first sentence vector, calculating the distance between the first sentence vector and each cluster center point in the N cluster center points, classifying the first sentence vector into a category corresponding to the cluster center point with the closest distance to the first sentence vector, wherein the first sentence vector is the sentence vector corresponding to any one of the remaining L-N standard problems, and L is the number of the standard problems; s4, after sentence vectors corresponding to all standard problems are classified, calculating cluster center points of N categories according to sentence vectors corresponding to the standard problems in each category, updating the cluster center points of the N categories, and executing S3 and S4 circularly until the distance between every two adjacent cluster center points of each category is within a preset distance.
The N value may be determined from a priori experience, which refers to the classification of all standard problems in the target database, so the N value may be determined from an existing classification. For example, assuming that n=30, calculating the distance between the sentence vector corresponding to the question to be queried and the cluster center point of each of the 30 cluster categories to obtain 30 distances, assuming that the smallest distance corresponds to the cluster center point of the 10 th cluster category, taking the 10 th cluster category as the target cluster category, assuming that the 10 th cluster category includes 50 standard questions (m=50), calculating the similarity between the question to be queried and the 50 standard questions in the target cluster category respectively to obtain 50 similarities, arranging the 50 similarities in descending order, sorting the similarity with the largest similarity among the 50 similarities, and taking the standard question corresponding to the similarity sorted in the first as the standard question with the highest similarity to be queried.
It should be noted that: the cluster center point of each of the N cluster categories is calculated in advance, the calculation is only needed once in advance, and the distance between the sentence vector corresponding to the to-be-queried problem and the cluster center point of each of the N cluster categories is calculated and only needed to be directly used. When the similarity is calculated, a clustering algorithm is adopted, so that the similarity between the to-be-queried problem and all standard problems in the target database is avoided, the calculated amount is greatly reduced, and the calculation efficiency is improved.
Optionally, calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category respectively includes: determining a first target word and a second target word, wherein the first target word is a word which appears in a to-be-queried problem and does not appear in a second standard problem, the second target word is a word which appears in the second standard problem and does not appear in the to-be-queried problem, and the second standard problem is any standard problem included in a target cluster class; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem and the first target word and the second target word. Optionally, calculating the similarity between the question to be queried and the second standard question according to the sentence vectors respectively corresponding to the question to be queried and the second standard question and the first target word and the second target word, including: calculating a similarity degree parameter between the first target word and the second target word; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem.
When calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category, the user inputs the inaccuracy of the description of the to-be-queried problem, so that the difference exists between the to-be-queried problem and the standard problem, but the semantics of the to-be-queried problem and the standard problem are basically consistent, so that the similarity between the to-be-queried problem and the standard problem needs to be calculated by considering not only the repeated words between the to-be-queried problem and the standard problem, but also the isolated words between the to-be-queried problem and the standard problem, and the similarity is calculated by considering the isolated words between the to-be-queried problem, wherein the first target word and the second target word are the isolated words, for example: the question to be queried is "do this can package the magic do? The second standard problem is "how does this beijing shanghai avoid the freight? The first target word, namely the solitary word in the to-be-queried problem, is "can", "package mail", "magic". When the similarity between the problem to be queried and each standard problem in the target cluster category is calculated, the contribution of the orphan between the problem to the similarity is considered, and the accuracy of a calculation result is high.
Optionally, calculating the similarity between the question to be queried and the second standard question according to the sentence vectors and the similarity parameters respectively corresponding to the question to be queried and the second standard question, including: according to the formula SIM (a, B) =Calculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Sentence vector corresponding to the second standard question,/>Modulo representing sentence vector corresponding to a question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
The formula for calculating the similarity degree parameter K of the first target word and the second target word is as follows: n is the number of first target words, ki is the semantic similarity parameter of the ith first target word, ki=w1×w2×s (1, 2), where W1 is the word weight of the ith first target word, W2 is the word weight of the second target word with the highest similarity to the ith first target word, and S (1, 2) is the semantic proximity parameter of the ith first target word and the second target word with the highest similarity to the ith first target word.
For example: the question to be queried is "do this can package the magic do? The second standard problem is "how does this beijing shanghai avoid the freight? The first target word, namely the solitary word in the question to be queried, is "can", "package mail", "magic" and "mock", the second target word, namely the solitary word in the second standard question, is "Beijing", "Shanghai", "free of freight" and "how", and the TF-IDF value of each word is searched from the corpus and is used as the word weight of the word and recorded as W, as shown in Table 1. TF-IDF (term frequency-inverse document frequency) is a common weighting technique for information retrieval and data mining. TF means word Frequency (Term Frequency), and IDF means inverse text Frequency index (Inverse Document Frequency). The more similar the words of two sentences, the more similar their contents should be. Thus, the degree of similarity of the word frequencies can be calculated starting from them.
TABLE 1
Based on word frequency, sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem are as follows:
Sentence vector corresponding to question to be queried
Sentence vector corresponding to the second standard problem
The number of the first target words is 4, and the first target words are respectively: can, package post, magic, mock, the word weight is respectively: w Can be used for ＝0.6、W Package post ＝7、W Magic all ＝5、W Does not take care of = 0.3.
The number of the second target words is 4, and the second target words are respectively: beijing, shanghai, no freight, and word weights are respectively: w Beijing ＝4.3、W Shanghai ＝4.8、W Free of freight ＝6、W How to = 1.
The corpus is queried to obtain semantic proximity parameters between a first target word and a second target word, as shown in table 2, the semantic proximity parameters between the first target word magic all and the second target word Shanghai are 1, and the semantic proximity parameters between the first target word magic all and the second target word Beijing are 0.2.
TABLE 2
| Beijing | Shanghai | Free of freight | How to | |
| Can be used for | 0 | 0 | 0 | 0 |
| Package post | 0 | 0 | 1 | 0 |
| Magic all | 0.2 | 1 | 0 | 0 |
| Does not take care of | 0 | 0 | 0 | 0.7 |
For the first target word "ok", i.e. 1 st first target word k1=0.
For the first target word "package mail", i.e., the 2 nd first target word, k2=7×6×1=42.
For the first target word "magic, i.e. the 3 rd first target word, k3=5×4.8×1=24.
For the first target word "mock", i.e. the 4 th first target word, k4=0.3×1×0.7=0.21.
The similarity parameter k=k1+k2+k3+k4=0+42+24+0.21= 66.21 between the first target word and the second target word.
According to the formulaCalculating the similarity between the question to be queried and the second standard question,/>And the similarity between the to-be-queried problem and the second standard problem is higher.
Fig. 2 is a schematic diagram of an alternative problem pushing apparatus for performing the problem pushing method according to an embodiment of the present invention, as shown in fig. 2, the apparatus includes: the device comprises an acquisition unit 10, a search unit 20 and a pushing unit 30.
An acquisition unit 10 for acquiring behavior data of the target user.
The searching unit 20 is configured to search for a standard problem corresponding to the behavior data of the target user according to a preset mapping relationship, so as to obtain a first standard problem, where the preset mapping relationship is a mapping relationship between the behavior data of the user and the standard problem stored in the target database.
And a pushing unit 30, configured to push the first standard problem to the target user.
Alternatively, the preset mapping relationship is determined in advance by the determining unit 40, and the determining unit 40 includes: the system comprises a receiving subunit, a collecting subunit, an extracting subunit, a screening subunit and a building subunit.
And the receiving subunit is used for receiving the problem inquiry requests sent by the plurality of users.
The acquisition subunit is used for acquiring behavior data generated when a plurality of users send problem inquiry requests, and the problem inquiry requests contain problems to be inquired.
And the extraction subunit is used for extracting the questions to be queried contained in the question query request.
And the screening subunit is used for screening out the standard problem with the highest similarity with the problem to be queried.
And the building subunit is used for building a mapping relation between the collected behavior data and the screened standard problems.
In the embodiment of the invention, the mapping relation between the behavior data of the user and the standard problem is determined in advance according to the behavior data of the historical user, and the behavior data of the target user is obtained; according to the mapping relation between the behavior data of the user and the standard questions stored in the target database, searching the standard questions corresponding to the behavior data of the target user to obtain a first standard question; the first standard questions are pushed to the target users, and the problems that the target users possibly ask questions are predicted according to the behavior data of the target users, so that the effect of intelligently recommending the problems to the users is achieved, and the problem that the intelligent customer service system in the prior art cannot intelligently recommend the problems to the users is solved.
Optionally, the screening subunit comprises: the device comprises a first calculation module, a screening module, a second calculation module and a determination module. The first calculation module is used for calculating the distance between the sentence vector corresponding to the problem to be queried and the cluster center point of each cluster type in the N cluster types to obtain N distances, wherein the N cluster types are determined in advance according to the following steps: acquiring all standard problems in a target database; all standard questions are respectively converted into corresponding sentence vectors; clustering sentence vectors corresponding to all standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2. And the screening module is used for screening the cluster category corresponding to the smallest distance in the N distances to obtain the target cluster category. The second calculation module is used for calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category respectively to obtain M similarity, wherein M is the number of standard problems in the target cluster category. And the determining module is used for taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried.
Optionally, the second computing module includes: the method comprises the steps of determining a sub-module and calculating a sub-module. The determining submodule is used for determining a first target word and a second target word, wherein the first target word is a word which appears in a to-be-queried problem and does not appear in a second standard problem, the second target word is a word which appears in the second standard problem and does not appear in the to-be-queried problem, and the second standard problem is any standard problem included in the target cluster category. And the calculation sub-module is used for calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem, as well as the first target word and the second target word.
Optionally, the calculating submodule includes: the system comprises a first large calculation module and a second large calculation module. The first calculation big module is used for calculating the similarity degree parameter between the first target word and the second target word. And the second calculation large module is used for calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors and the similarity degree parameters respectively corresponding to the to-be-queried problem and the second standard problem.
Optionally, the second calculation large module includes: and calculating a small module. A calculation small module for calculating the formulaCalculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Sentence vector corresponding to the second standard question,/>Modulo representing sentence vector corresponding to a question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
In one aspect, an embodiment of the present invention provides a storage medium, where the storage medium includes a stored program, and when the program runs, controls a device where the storage medium is located to execute the following steps: acquiring behavior data of a target user; searching a standard problem corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problem stored in the target database, and the preset mapping relation is determined in advance through the following steps: receiving a problem inquiry request sent by a plurality of users, and collecting behavior data generated when the plurality of users send the problem inquiry request, wherein the problem inquiry request comprises a problem to be inquired; extracting the problems to be queried contained in the problem query request, and screening out the standard problems with highest similarity with the problems to be queried; establishing a mapping relation between the collected behavior data and the screened standard problems; the first criteria question is pushed to the target user.
Optionally, the device controlling the storage medium when the program runs further performs the following steps: calculating distances between sentence vectors corresponding to the to-be-queried problem and cluster center points of each of N cluster categories to obtain N distances, wherein the N cluster categories are determined in advance according to the following steps: acquiring all standard problems in a target database; all standard questions are respectively converted into corresponding sentence vectors; clustering sentence vectors corresponding to all standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2; screening out the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category; respectively calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category to obtain M similarity, wherein M is the number of standard problems in the target cluster category; and taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried.
Optionally, the device controlling the storage medium when the program runs further performs the following steps: determining a first target word and a second target word, wherein the first target word is a word which appears in a to-be-queried problem and does not appear in a second standard problem, the second target word is a word which appears in the second standard problem and does not appear in the to-be-queried problem, and the second standard problem is any standard problem included in a target cluster class; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem and the first target word and the second target word.
Optionally, the device controlling the storage medium when the program runs further performs the following steps: calculating a similarity degree parameter between the first target word and the second target word; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem.
Optionally, the device controlling the storage medium when the program runs further performs the following steps: according to the formulaCalculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Sentence vector corresponding to the second standard question,/>Modulo representing sentence vector corresponding to a question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
In one aspect, an embodiment of the present invention provides a computer device, including a memory for storing information including program instructions, and a processor for controlling execution of the program instructions, the program instructions when loaded and executed by the processor implementing the steps of: acquiring behavior data of a target user; searching a standard problem corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problem stored in the target database, and the preset mapping relation is determined in advance through the following steps: receiving a problem inquiry request sent by a plurality of users, and collecting behavior data generated when the plurality of users send the problem inquiry request, wherein the problem inquiry request comprises a problem to be inquired; extracting the problems to be queried contained in the problem query request, and screening out the standard problems with highest similarity with the problems to be queried; establishing a mapping relation between the collected behavior data and the screened standard problems; the first criteria question is pushed to the target user.
Optionally, the program instructions when loaded and executed by the processor further implement the steps of: calculating distances between sentence vectors corresponding to the to-be-queried problem and cluster center points of each of N cluster categories to obtain N distances, wherein the N cluster categories are determined in advance according to the following steps: acquiring all standard problems in a target database; all standard questions are respectively converted into corresponding sentence vectors; clustering sentence vectors corresponding to all standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2; screening out the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category; respectively calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category to obtain M similarity, wherein M is the number of standard problems in the target cluster category; and taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried.
Optionally, the program instructions when loaded and executed by the processor further implement the steps of: determining a first target word and a second target word, wherein the first target word is a word which appears in a to-be-queried problem and does not appear in a second standard problem, the second target word is a word which appears in the second standard problem and does not appear in the to-be-queried problem, and the second standard problem is any standard problem included in a target cluster class; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem and the first target word and the second target word.
Optionally, the program instructions when loaded and executed by the processor further implement the steps of: calculating a similarity degree parameter between the first target word and the second target word; and calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem.
Optionally, the program instructions when loaded and executed by the processor further implement the steps of: according to the formulaCalculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Sentence vector corresponding to the second standard question,/>Modulo representing sentence vector corresponding to a question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
Fig. 3 is a schematic diagram of a computer device according to an embodiment of the present invention. As shown in fig. 3, the computer device 50 of this embodiment includes: the processor 51, the memory 52, and the computer program 53 stored in the memory 52 and capable of running on the processor 51, the computer program 53 when executed by the processor 51 implements the problem pushing method in the embodiment, and is not described herein in detail to avoid repetition. Or the computer program, when executed by the processor 51, implements the functions of each model/unit in the problem pushing device in the embodiment, and in order to avoid repetition, it is not described in detail herein.
The computer device 50 may be a desktop computer, a notebook computer, a palm top computer, a cloud server, or the like. Computer devices may include, but are not limited to, a processor 51, a memory 52. It will be appreciated by those skilled in the art that fig. 3 is merely an example of computer device 50 and is not intended to limit computer device 50, and may include more or fewer components than shown, or may combine certain components, or different components, e.g., a computer device may also include an input-output device, a network access device, a bus, etc.
The Processor 51 may be a central processing unit (Central Processing Unit, CPU), other general purpose Processor, digital signal Processor (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), field-Programmable gate array (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 52 may be an internal storage unit of the computer device 50, such as a hard disk or memory of the computer device 50. The memory 52 may also be an external storage device of the computer device 50, such as a plug-in hard disk provided on the computer device 50, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD), or the like. Further, the memory 52 may also include both internal storage units and external storage devices of the computer device 50. The memory 52 is used to store computer programs and other programs and data required by the computer device. The memory 52 may also be used to temporarily store data that has been output or is to be output.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided in the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the elements is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in hardware plus software functional units.
The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a Processor (Processor) to perform part of the steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather to enable any modification, equivalent replacement, improvement or the like to be made within the spirit and principles of the invention.
Claims (4)
1. A problem pushing method, characterized in that the method comprises:
acquiring behavior data of a target user;
Searching a standard problem corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problem stored in a target database, and the preset mapping relation is determined in advance through the following steps: receiving a problem inquiry request sent by a plurality of users, and collecting behavior data generated when the plurality of users send the problem inquiry request, wherein the problem inquiry request comprises a problem to be inquired; extracting the problems to be queried contained in the problem query request, and screening out the standard problems with highest similarity with the problems to be queried; establishing a mapping relation between the collected behavior data and the screened standard problems;
pushing the first standard question to the target user;
The screening out the standard problem with the highest similarity with the problem to be queried comprises the following steps:
Calculating the distance between the sentence vector corresponding to the problem to be queried and the cluster center point of each cluster category in N cluster categories to obtain N distances, wherein the N cluster categories are determined in advance according to the following steps: acquiring all standard problems in the target database; respectively converting all the standard questions into corresponding sentence vectors; clustering sentence vectors corresponding to all the standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2;
Screening out the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category;
respectively calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category to obtain M similarity, wherein M is the number of standard problems in the target cluster category;
Taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried;
The calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category respectively includes:
Determining a first target word and a second target word, wherein the first target word is a word which appears in the to-be-queried problem and does not appear in a second standard problem, the second target word is a word which appears in the second standard problem and does not appear in the to-be-queried problem, and the second standard problem is any standard problem included in the target cluster category;
Calculating the similarity between the to-be-queried problem and the second standard problem according to sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem, the first target word and the second target word;
The calculating the similarity between the query question and the second standard question according to the sentence vectors respectively corresponding to the query question and the second standard question and the first target word and the second target word includes:
calculating a similarity degree parameter between the first target word and the second target word;
calculating the similarity between the to-be-queried problem and the second standard problem according to sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem;
the calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem includes:
according to the formula Calculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Representing sentence vector corresponding to the second standard question,/>Modulus of sentence vector corresponding to the question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
2. A problem pushing device, the device comprising:
the acquisition unit is used for acquiring behavior data of the target user;
The searching unit is used for searching standard problems corresponding to the behavior data of the target user according to a preset mapping relation to obtain a first standard problem, wherein the preset mapping relation is a mapping relation between the behavior data of the user and the standard problems stored in the target database;
A pushing unit for pushing the first standard question to the target user,
Wherein the preset mapping relationship is determined in advance by a determining unit, the determining unit includes:
A receiving subunit, configured to receive question query requests sent by a plurality of users;
The acquisition subunit is used for acquiring behavior data generated when the plurality of users send the problem inquiry requests, wherein the problem inquiry requests comprise problems to be inquired;
An extracting subunit, configured to extract a problem to be queried included in the problem query request;
the screening subunit is used for screening out the standard problem with highest similarity with the problem to be queried;
The building subunit is used for building a mapping relation between the collected behavior data and the screened standard problems;
The screening subunit comprises:
the first calculation module is used for calculating the distance between the sentence vector corresponding to the to-be-queried problem and the cluster center point of each cluster type in the N cluster types to obtain N distances, wherein the N cluster types are determined in advance according to the following steps: acquiring all standard problems in the target database; respectively converting all the standard questions into corresponding sentence vectors; clustering sentence vectors corresponding to all the standard problems to obtain N clustering categories, wherein N is a natural number greater than or equal to 2;
The screening module is used for screening the cluster category corresponding to the smallest distance in the N distances to obtain a target cluster category;
The second calculation module is used for calculating the similarity between the to-be-queried problem and each standard problem in the target cluster category respectively to obtain M similarity, wherein M is the number of standard problems in the target cluster category;
the determining module is used for taking the standard problem corresponding to the maximum similarity in the M similarity as the standard problem with the highest similarity with the problem to be queried;
The second computing module includes:
A determining submodule, configured to determine a first target term and a second target term, where the first target term is a term that appears in the question to be queried and does not appear in a second standard question, the second target term is a term that appears in the second standard question and does not appear in the question to be queried, and the second standard question is any standard question included in the target cluster category;
The calculation sub-module is used for calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors respectively corresponding to the to-be-queried problem and the second standard problem, the first target word and the second target word;
The calculating the similarity between the query question and the second standard question according to the sentence vectors respectively corresponding to the query question and the second standard question and the first target word and the second target word includes:
calculating a similarity degree parameter between the first target word and the second target word;
calculating the similarity between the to-be-queried problem and the second standard problem according to sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem;
the calculating the similarity between the to-be-queried problem and the second standard problem according to the sentence vectors and the similarity parameters respectively corresponding to the to-be-queried problem and the second standard problem includes:
according to the formula Calculating the similarity between the question to be queried and the second standard question, wherein SIM (A, B) represents the similarity between the question to be queried and the second standard question,/>Representing sentence vector corresponding to the question to be queried,/>Representing sentence vector corresponding to the second standard question,/>Modulus of sentence vector corresponding to the question to be queried,/>And K represents a similarity degree parameter between the first target word and the second target word.
3. A storage medium comprising a stored program, wherein the program, when run, controls a device in which the storage medium is located to perform the problem pushing method of claim 1.
4. A computer device comprising a memory for storing information including program instructions and a processor for controlling execution of the program instructions, characterized by: which when loaded and executed by a processor carries out the steps of the problem pushing method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910038458.9A CN109885651B (en) | 2019-01-16 | 2019-01-16 | Question pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910038458.9A CN109885651B (en) | 2019-01-16 | 2019-01-16 | Question pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109885651A CN109885651A (en) | 2019-06-14 |
| CN109885651B true CN109885651B (en) | 2024-06-04 |
Family
ID=66926094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910038458.9A Active CN109885651B (en) | 2019-01-16 | 2019-01-16 | Question pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109885651B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111143530B (en) * | 2019-12-24 | 2024-04-05 | 平安健康保险股份有限公司 | Intelligent reply method and device |
| CN113505293B (en) * | 2021-06-15 | 2024-03-19 | 深圳追一科技有限公司 | Information pushing method and device, electronic equipment and storage medium |
| CN113407700A (en) * | 2021-07-06 | 2021-09-17 | 中国工商银行股份有限公司 | Data query method, device and equipment |
| CN114840762B (en) * | 2022-05-19 | 2024-08-23 | 马上消费金融股份有限公司 | Recommended content determining method and device and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104184705A (en) * | 2013-05-23 | 2014-12-03 | 腾讯科技（深圳）有限公司 | Verification method, apparatus, server, user data center and system |
| CN106649742A (en) * | 2016-12-26 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Database maintenance method and device |
| CN106776751A (en) * | 2016-11-22 | 2017-05-31 | 上海智臻智能网络科技股份有限公司 | The clustering method and clustering apparatus of a kind of data |
| CN106897334A (en) * | 2016-06-24 | 2017-06-27 | 阿里巴巴集团控股有限公司 | A kind of question pushing method and equipment |
| CN108268877A (en) * | 2016-12-30 | 2018-07-10 | 中国移动通信集团黑龙江有限公司 | A kind of method and apparatus for identifying target terminal |
| CN108804567A (en) * | 2018-05-22 | 2018-11-13 | 平安科技（深圳）有限公司 | Method, equipment, storage medium and device for improving intelligent customer service response rate |
| CN109033156A (en) * | 2018-06-13 | 2018-12-18 | 腾讯科技（深圳）有限公司 | A kind of information processing method, device and terminal |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107220380A (en) * | 2017-06-27 | 2017-09-29 | 北京百度网讯科技有限公司 | Question and answer based on artificial intelligence recommend method, device and computer equipment |
-
2019
- 2019-01-16 CN CN201910038458.9A patent/CN109885651B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104184705A (en) * | 2013-05-23 | 2014-12-03 | 腾讯科技（深圳）有限公司 | Verification method, apparatus, server, user data center and system |
| CN106897334A (en) * | 2016-06-24 | 2017-06-27 | 阿里巴巴集团控股有限公司 | A kind of question pushing method and equipment |
| CN106776751A (en) * | 2016-11-22 | 2017-05-31 | 上海智臻智能网络科技股份有限公司 | The clustering method and clustering apparatus of a kind of data |
| CN106649742A (en) * | 2016-12-26 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Database maintenance method and device |
| CN108268877A (en) * | 2016-12-30 | 2018-07-10 | 中国移动通信集团黑龙江有限公司 | A kind of method and apparatus for identifying target terminal |
| CN108804567A (en) * | 2018-05-22 | 2018-11-13 | 平安科技（深圳）有限公司 | Method, equipment, storage medium and device for improving intelligent customer service response rate |
| CN109033156A (en) * | 2018-06-13 | 2018-12-18 | 腾讯科技（深圳）有限公司 | A kind of information processing method, device and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109885651A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109885651B (en) | Question pushing method and device | |
| CN109885773B (en) | Personalized article recommendation method, system, medium and equipment | |
| CN112035599B (en) | Query method and device based on vertical search, computer equipment and storage medium | |
| EP3540612A1 (en) | Cluster processing method and device for questions in automatic question and answering system | |
| WO2017097231A1 (en) | Topic processing method and device | |
| US10482146B2 (en) | Systems and methods for automatic customization of content filtering | |
| CN109816482B (en) | Knowledge graph construction method, device and equipment of e-commerce platform and storage medium | |
| CN111767716B (en) | Method and device for determining enterprise multi-level industry information and computer equipment | |
| CN104834651B (en) | Method and device for providing high-frequency question answers | |
| CN108804642A (en) | Search method, device, computer equipment and storage medium | |
| JP6355840B2 (en) | Stopword identification method and apparatus | |
| CN103136228A (en) | Image search method and image search device | |
| US20110191335A1 (en) | Method and system for conducting legal research using clustering analytics | |
| CN111460090A (en) | Vector-based document retrieval method and device, computer equipment and storage medium | |
| CN110309377B (en) | Semantic normalization, question pattern generation and response determination methods and devices | |
| CN104537341A (en) | Human face picture information obtaining method and device | |
| CN108875065B (en) | Indonesia news webpage recommendation method based on content | |
| US9552415B2 (en) | Category classification processing device and method | |
| CN110245289A (en) | A kind of information search method and relevant device | |
| CN112100396A (en) | Data processing method and device | |
| CN110674365A (en) | Searching method, device, equipment and storage medium | |
| CN112632261A (en) | Intelligent question and answer method, device, equipment and storage medium | |
| CN117539990A (en) | Problem processing method and device, electronic equipment and storage medium | |
| CN111177372A (en) | Scientific and technological achievement classification method, device, equipment and medium | |
| CN109960719A (en) | A kind of document handling method and relevant 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 |