CN115878780A - Question pushing method and device, computer equipment and storage medium - Google Patents
Question pushing method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN115878780A CN115878780A CN202211499888.9A CN202211499888A CN115878780A CN 115878780 A CN115878780 A CN 115878780A CN 202211499888 A CN202211499888 A CN 202211499888A CN 115878780 A CN115878780 A CN 115878780A
- Authority
- CN
- China
- Prior art keywords
- question
- user
- data
- target
- candidate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000003993 interaction Effects 0.000 claims abstract description 102
- 238000012216 screening Methods 0.000 claims abstract description 14
- 230000006399 behavior Effects 0.000 claims description 43
- 230000004044 response Effects 0.000 claims description 33
- 230000001960 triggered effect Effects 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 13
- 238000005065 mining Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000001502 supplementing effect Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims 1
- 238000004891 communication Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000002452 interceptive effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007711 solidification Methods 0.000 description 1
- 230000008023 solidification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000000699 topical effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to a question pushing method, a question pushing device, computer equipment and a storage medium. The method comprises the following steps: acquiring user tag data, user behavior data, historical interaction data and service scene characteristics corresponding to the current position of a user; establishing a baffle rule according to the user label data, the user behavior data and the service scene characteristics; screening original problems in a preset first problem library according to the baffle rule to obtain a first candidate problem set; determining a target keyword according to the user tag data, the historical interaction data and the user behavior data; recalling the first question bank according to the target keywords to obtain a second candidate question set; and determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set, and pushing the first question sequence to a display interface corresponding to the user. By adopting the method, the problem pushing accuracy can be improved.
Description
Technical Field
The present application relates to the field of artificial intelligence technologies, and in particular, to a problem pushing method and apparatus, a computer device, and a storage medium.
Background
With the rapid development of artificial intelligence technology, problem intelligence push technology has emerged. The intelligent problem pushing can refer to a service which can know the appeal faced by the user and recommend the problem which the user wants to consult according to the appeal of the user when the user establishes the chat session.
In the related technology, hot problems frequently consulted by a client are directly selected from a background customer service system and recommended to the user. However, different clients often have different consultation problems, and the problem push is easily inaccurate due to the push mode of directly pushing the hotspot problem. Therefore, how to improve the accuracy of problem pushing becomes a technical problem that needs to be solved urgently by those skilled in the art.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a problem pushing method, apparatus, computer device and storage medium capable of improving problem pushing accuracy.
In a first aspect, the present application provides a problem pushing method. The method comprises the following steps:
acquiring user tag data, user behavior data, historical interaction data and service scene characteristics corresponding to the current position of a user;
establishing a baffle rule according to the user label data, the user behavior data and the service scene characteristics;
screening original problems in a preset first problem library according to the baffle rule to obtain a first candidate problem set;
determining a target keyword according to the user tag data, the historical interaction data and the user behavior data;
recalling the first question bank according to the target keywords to obtain a second candidate question set;
and determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set, and pushing the first question sequence to a display interface corresponding to the user.
In one embodiment, the method further comprises:
determining a target question selected by the user in response to a first operation triggered on a display interface;
carrying out classification matching on the target problems according to a preset classification problem library to obtain a classification matching result;
and verifying the user according to the classification matching result, or determining a question response corresponding to the target question based on the user attribute information of the user and a preset standard response library.
In one embodiment, the determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set includes:
obtaining a target question set according to the first candidate question set and the second candidate question set;
sequencing the target problem set, and determining a plurality of first problems from the sequenced target problem set;
determining a plurality of second questions from a second question bank according to the historical interaction data;
and obtaining the first question sequence according to the combination of the plurality of first questions and the plurality of second questions.
In one embodiment, the method further comprises:
in response to a second operation triggered on a display interface or an operation not triggered by the display interface in a preset time period, removing the first questions from the target question set and removing the second questions from a second question bank;
determining a plurality of third questions from the removed target question set, and determining a plurality of fourth questions from the removed second question bank;
combining the plurality of third questions and the plurality of fourth questions to obtain a second question sequence;
and pushing the second question sequence to the display interface.
In one embodiment, before the obtaining of the user tag data, the user behavior data, the historical interaction data, and the service scenario features corresponding to the current location of the user, the method further includes:
acquiring operation information triggered by the user on a customer service system and response information generated by the customer service system according to the operation information;
determining the user behavior data based on the operation information and the response information.
In one embodiment, the method further comprises:
acquiring historical interaction information of a user;
converting the historical interaction information into an interaction vector according to a preset pre-training model;
mining according to the interaction vector to obtain a problem cluster; the problem cluster comprises at least one problem;
and matching each question in the question cluster according to the first question bank, and supplementing the matched question cluster into the first question bank.
In one embodiment, the converting the historical interaction information into an interaction vector according to a preset pre-training model includes:
the historical interaction information is cleaned and removed from the weight, and target interaction information is obtained;
and encoding the target interaction information into the interaction vector according to the pre-training model.
In a second aspect, the present application further provides a problem pushing device. The device comprises:
the data acquisition module is used for acquiring user tag data, user behavior data, historical interaction data of a user and service scene characteristics corresponding to the current position of the user;
the rule establishing module is used for establishing a baffle rule according to the user label data, the user behavior data and the service scene characteristics;
the screening module is used for screening original problems in a preset first problem library according to the baffle rule to obtain a first candidate problem set;
the target keyword determining module is used for determining a target keyword according to the user tag data, the historical interaction data and the user behavior data;
the recall module is used for recalling the first question bank according to the target keywords to obtain a second candidate question set;
and the pushing module is used for determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set and pushing the first question sequence to a display interface corresponding to the user.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the problem pushing method when executing the computer program.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the problem pushing method described above.
In a fifth aspect, the present application further provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, performs the steps of the problem pushing method described above.
According to the problem pushing method, the device, the computer equipment and the storage medium, the user label data, the user behavior data, the historical interaction data and the service scene characteristics corresponding to the current position of the user are obtained, then the baffle rule is established according to the user label data, the user behavior data and the service scene characteristics, so that the baffle rule is highly related to the user label data, the user behavior data, the historical interaction data and the service scene characteristics of the current position of the user, the relevance between the first candidate problem set obtained by subsequent screening and the problem which the user wants to consult is improved, the problem pushing accuracy is improved, the second candidate problem is obtained by recalling the first problem base according to the target keyword, the first problem sequence is determined according to the historical interaction data, the first candidate problem set and the second candidate problem set, the first problem sequence is pushed to the user, the problem which the user wants to consult is convenient to select, and the problem pushing accuracy is further improved.
Drawings
FIG. 1 is a diagram of an application environment of a problem pushing method in one embodiment;
FIG. 2 is a first flowchart of a problem pushing method according to an embodiment;
FIG. 3 is a second flowchart of a problem pushing method according to an embodiment;
FIG. 4 is a flowchart illustrating the steps of combining to obtain a first question sequence in one embodiment;
FIG. 5 is a third flowchart of a problem pushing method according to an embodiment;
FIG. 6 is a fourth flowchart illustrating a method for pushing questions in one embodiment;
FIG. 7 is a block diagram of an issue pushing device in one embodiment;
FIG. 8 is a diagram of an internal structure of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The problem pushing method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The server acquires user tag data of a user, user behavior data of the user on the terminal 102, historical interaction data of the user on the terminal 102 and service scene characteristics corresponding to the current position of the user; establishing a baffle rule according to the user tag data, the user behavior data and the service scene characteristics, screening original problems in a preset first problem bank according to the baffle rule to obtain a first candidate problem set, determining target keywords according to the user tag data, the historical interaction data and the user behavior data, and recalling the first problem bank according to the target keywords to obtain a second candidate problem set; and determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set, and pushing the first question sequence to a display interface corresponding to the user. The terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, and tablet computers. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers.
In one embodiment, as shown in fig. 2, a question pushing method is provided, which is described by taking the method as an example applied to the server 104 in fig. 1, and includes the following steps:
Wherein, the user tag data can refer to the tag data of the user in the server and in the terminal. The user tag data may include, but is not limited to, the account status of the user, credit information of the user, identity information of the user, and the like. The account status includes normal status, blacklisted users, governed users, invalid users, low credit users, high credit users, etc. The amount information may refer to the amount of funds that the user is able to debit. Identity information may refer to information that characterizes a unique identity of a user. The identity information can be represented by the identity card number, the mobile phone number and the like of the user.
The user behavior data may refer to data of a behavior performed by a user in the terminal. The user behavior data may include, but is not limited to, operations performed, actions taken, etc. by the user in the terminal.
In some embodiments, the user behavior data may also include responses generated by the terminal in accordance with operations performed by the user.
The service scenario features may characterize the service scenario in which the user is currently located in the terminal. For example, the business scenario may include, but is not limited to, a mall, a debit, a payment, a customer service finding, and the like. The service scene features are features that are not specific to the user himself.
The historical interaction data may refer to historical data consulted by the user in the customer service system and interaction data between the terminal and the user. The historical interaction data may include, but is not limited to, historical interaction text, interaction speech, and the like.
Illustratively, the user tag data, the user behavior data, the historical interaction data, and the like may be stored in a server, and the server obtains the user tag data, the user behavior data, and the historical interaction data directly from the invocation. And the server determines the service scene characteristics according to the position of the current user fed back by the terminal.
And step 204, establishing a baffle rule according to the user label data, the user behavior data and the service scene characteristics.
Wherein, the blind rule may refer to a rule for screening the original problem. The baffle rule is formed by the characteristics of the user and the characteristics which are not specific to the user.
Illustratively, the service scene where the current user is located is determined according to the service scene characteristics, then the operation which is performed by the current user is determined according to the user behavior data, the problem encountered by the user is determined according to the user tag data, and then the baffle rule is generated based on the service scene of the user, the operation which is performed by the user and the problem encountered by the user.
And step 206, screening original problems in a preset first problem library according to the baffle rule to obtain a first candidate problem set.
The first question bank is preset by the server, and a plurality of original questions are set in the first question bank.
Illustratively, according to the baffle rule established in the previous step, the original questions in the first question bank are screened to obtain a first candidate question set meeting the baffle rule.
Illustratively, a mapping relationship may be established with the first question bank according to the blind rule, and then the first candidate question set may be determined based on the mapping relationship.
For example, when the user is an unregistered client, the first candidate question set may be a question type belonging to account registration and introduction of related services and related functions.
And step 208, determining the target keywords according to the user tag data, the historical interaction data and the user behavior data.
Illustratively, the user tag data, the historical interaction data and the user behavior data may be extracted to obtain the target keyword.
And step 210, recalling the first question bank according to the target keywords to obtain a second candidate question set.
For example, the target keyword may be input into a recall model to recall the first question bank to obtain a second candidate question set.
For example, the BM25 algorithm is adopted to recall the first question bank according to the target keyword to find a second candidate question set with a higher similarity to the target keyword text, and then the second candidate question set is ranked according to the target keyword to obtain a ranked second candidate question set.
And step 212, determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set, and pushing the first question sequence to a display interface corresponding to the user.
The first question sequence may refer to a question sequence which is pushed to a user and displayed on a display interface corresponding to a terminal used by the user. The first question sequence may include a plurality of questions.
The display interface may refer to a display interface corresponding to a terminal used by a user.
Illustratively, a number of questions may be selected from the first candidate question set and the second candidate question set, then a number of questions may be selected from the hot question bank according to the historical interaction data, the selected questions may be combined to obtain a first question sequence, and the first question sequence may be pushed to a display interface corresponding to the user.
According to the problem pushing method, the user label data, the user behavior data, the historical interaction data and the service scene characteristics corresponding to the current position of the user are obtained, the baffle rule is established according to the user label data, the user behavior data and the service scene characteristics, so that the baffle rule is highly related to the user label data, the user behavior data, the historical interaction data and the service scene characteristics of the current position of the user, the relevance of a first candidate problem set obtained through subsequent screening and a problem which the user wants to consult is improved, the problem pushing accuracy is improved, a second candidate problem is obtained by recalling a first problem base according to a target keyword, a first problem sequence is determined according to the historical interaction data, the first candidate problem set and the second candidate problem set, the first problem sequence is pushed to the user, the user can conveniently select the problem which the user wants to consult, and the problem pushing accuracy is further improved.
Referring to fig. 3, in some embodiments, the question pushing method further includes, but is not limited to, the following steps:
The first operation may refer to a question selection operation triggered by a user on the display interface. For example, the user clicks on the first question sequence on the display interface displaying the first question sequence.
Illustratively, the target question selected by the user is determined according to the operation of clicking the first question sequence on the display interface displaying the first question sequence.
And step 304, carrying out classification matching on the target questions according to a preset classification question library to obtain a classification matching result.
The classification question bank may refer to a question set used for performing classification matching processing on questions. The classification problem library is preset and can be preset in the server by an administrator.
The classification matching result may include, but is not limited to, a first classification result for characterizing that the target question belongs to the classification question bank and a second classification result for characterizing that the target question does not belong to the classification question bank.
Illustratively, the target question is subjected to classification matching processing according to the classification question bank to determine whether the target question belongs to a question in the classification question bank, so as to obtain a classification result.
And step 306, verifying the user according to the classification matching result, or determining a question response corresponding to the target question based on the user attribute information of the user and a preset standard response library.
The user attribute information may refer to basic attribute information of the user. The user attribute information may include gender information of the user, age information of the user, and the like.
The standard answer library may refer to a set of answers corresponding to the question. The standard answer library may correspond to the first question library and the second question library, and may include answers corresponding to all questions in the first question library and the second question library.
Illustratively, the user is verified according to the classification matching result, or a question response corresponding to the target question is determined based on the user attribute information of the user and a preset standard response library.
Illustratively, the user is authenticated if the classification matching result is the first classification result characterizing that the target question belongs to the classification question bank. And if the classification result is a second classification result which represents that the target question does not belong to the classification question library, determining a question response corresponding to the target question according to the user attribute information and the standard response library.
For example, when the user is a male in the third decade of life and the target question is "how to perform account registration", the target question does not belong to a question in the category question bank, and the user attribute information includes the third decade of life and the male, the answer to the question obtained according to the user attribute information and the characterization answer bank may be "mr. You are good, please perform account registration according to the following directions".
For example, when the target question is "how to borrow", security verification is required for the user, including but not limited to sending an identity authentication request to the user, sending a password verification to the user, and the like.
Referring to FIG. 4, in some embodiments, step 212 includes, but is not limited to, the following steps:
Wherein the target problem set may refer to a set of problems of the first candidate problem set and the second candidate problem set.
Illustratively, the first candidate question set and the second candidate question set may be combined to obtain a target question set.
Illustratively, the questions in the target question set may be sorted according to the target keywords, resulting in a sorted target question set, and a plurality of first questions may be selected in the sorted target question set.
For example, the text similarity or relevance between the target keyword and the questions in the target question set may be calculated, then the target question set may be obtained by sorting the target keyword and the questions in the target question set from high to low according to the text similarity or relevance, and then a plurality of first questions may be selected from the sorted target question set. Such as the first question to select the top third of the ranking.
Wherein, the second question bank is a preset question bank. The second question bank may be a hot question bank, which may be preconfigured in the server by the administrator.
Illustratively, a plurality of second questions may be selected from a hot question bank based on historical interaction data.
For example, a plurality of second questions may be obtained by selecting similar or identical questions from the topical question library based on the questions consulted by the user in the historical interaction data.
Illustratively, the plurality of first questions and the plurality of second questions obtained in the previous steps are combined to obtain a first question sequence.
According to the technical scheme, the plurality of first problems are selected from the first problem base, the plurality of second problems are selected from the second problem base, the first problems and the second problems are combined to obtain the first problem sequence, the problem pushing accuracy can be improved, the first problem sequence is a sequence after sequencing is carried out, therefore, a user can conveniently select a target problem, and the user experience is improved.
In some embodiments, please refer to fig. 5, in some embodiments, the question pushing method further includes, but is not limited to, the following steps:
step 502, in response to a second operation triggered on the display interface or an operation not triggered in a preset time period on the display interface, removing a plurality of first questions from the target question set and removing a plurality of second questions from the second question bank.
The second operation may refer to a selection operation triggered by the user on the display interface. The second operation may be that the user clicks a "none at all" button on the display interface, or the like, or may be that the user clicks a "next batch" operation on the display interface, or the like.
Illustratively, when the user triggers the second operation on the display interface, or the display interface does not trigger the operation within a preset time period, the situation is that the first question sequence for pushing is not satisfied, or a question that the user wants to consult does not exist in the first question sequence. In this case, the server removes the first questions from the target set of questions and removes the second questions from the second question bank in order to push the questions to the user again.
It should be noted that the second question is not completely deleted from the second question bank, but is not recommended when the questions are pushed to the same user.
Illustratively, the plurality of third questions and the plurality of fourth questions determined in the foregoing steps are combined to obtain the second question sequence.
And step 508, pushing the second question sequence to a display interface.
Illustratively, the second question sequence is pushed to a display interface of a terminal used by the same user, so that the user can select the questions in the second question sequence conveniently.
It should be noted that, for the target question selected in the second question sequence, it is also necessary to perform classification matching processing on the target question according to the classification question library to obtain a classification matching result, and then verify the user according to the classification matching result, or determine the question answer of the target question based on the user attribute information and the standard answer library.
According to the technical scheme of the embodiment of the application, when the first problem sequence does not meet the problem that the client wants to consult, the second problem sequence can be pushed again, and therefore the problem pushing accuracy is improved.
In some embodiments, the question pushing method further comprises, but is not limited to, the steps of: and recording the problem pushing times, if the problem pushing times are larger than the pushing time threshold value, not performing problem pushing, and sending a switching request to connect the manual customer service.
According to the technical scheme, when the problem pushing times are larger than the pushing time threshold value, the problem pushing is not carried out any more, manual customer service is switched, and the experience of a user can be improved.
In some embodiments, the question pushing method further includes, but is not limited to, the steps of: acquiring operation information triggered by a user on a customer service system and response information generated by the customer service system according to the operation information; user behavior data is determined based on the operational information and the response information.
The operation information may refer to an operation performed by a user on the customer service system. The customer service system can be a system for providing customer service consultation for users, and the customer service system can be embedded in the terminal and is in communication connection with the server. The server may be provided, and the terminal may be provided with a customer service entrance.
The response information may refer to a reply and a response information generated by the customer service system with respect to the operation of the user.
Illustratively, the server acquires the operation triggered by the user on the customer service system and the response information generated by the customer service system to the operation information, and then determines the user behavior data of the user according to the operation information and the response information.
Referring to fig. 6, in some embodiments, the question pushing method further includes, but is not limited to, the following steps:
The historical interaction information can refer to interaction information between a user on the terminal and the customer service system. The historical interaction information may include the aforementioned historical interaction data and current interaction information for the user and the customer service system.
Illustratively, the server acquires current interaction information of the user and the customer service system, extracts historical interaction data of the user, and acquires the historical interaction information according to the historical interaction data and the current interaction information.
And step 604, converting the historical interaction information into an interaction vector according to a preset pre-training model.
The pre-training model may refer to a model for converting the historical interaction information. The pre-training model may be a Roformer-sim model. The RoFormer-Sim model is a model integrating retrieval and generation.
Illustratively, the historical interaction information may be input to a Roformer-sim model for processing to obtain an interaction vector.
Illustratively, a problem cluster can be obtained by mining according to an interaction vector by adopting methods such as a connected subgraph and a free solidification degree.
Illustratively, the text similarity of the original question in the first question bank and each question in the question cluster can be calculated, and the question in the question cluster with the text similarity larger than the similarity threshold value is supplemented into the first question bank.
Illustratively, the questions in the question cluster may also be clustered and text generated to obtain a batch of new questions, which are then added to the first question bank.
In some embodiments, step 604 includes, but is not limited to, the following steps: the historical interaction information is cleaned and removed from the weight, and target interaction information is obtained; and encoding the target interaction information into an interaction vector according to the pre-training model.
The operation of removing the duplicate may refer to deleting sensitive information, symbols, keywords, and the like in the information. The sensitive information includes identity information, passwords, etc. of the user.
Illustratively, firstly, sensitive information, symbols, keywords and the like in the historical interaction information are cleaned and removed, so that target interaction information is obtained, and the target interaction information is input into a pre-training model to be encoded, so that an interaction vector is obtained.
According to the technical scheme, the historical interactive information is cleaned to remove the repeated information, so that the sensitive information of the user is protected conveniently, and the safety of the user data is improved.
It should be understood that, although the steps in the flowcharts related to the embodiments are shown in sequence as indicated by the arrows, the steps are not necessarily executed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the above embodiments may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the present application further provides a problem pushing device for implementing the problem pushing method mentioned above. The implementation of the solution provided by the device is similar to the implementation described in the above method.
In one embodiment, as shown in fig. 7, there is provided a question pushing apparatus including: a data acquisition module 702, a rule establishment module 704, a screening module 706, a target keyword determination module 708, a recall module 710, and a push module 712, wherein:
the data obtaining module 702 is configured to obtain user tag data, user behavior data, historical interaction data, and a service scene characteristic corresponding to a current location of the user.
And a rule establishing module 704, configured to establish a blind rule according to the user tag data, the user behavior data, and the service scene characteristic.
The screening module 706 is configured to screen original questions in a preset first question bank according to the baffle rule to obtain a first candidate question set.
And a target keyword determination module 708, configured to determine a target keyword according to the user tag data, the historical interaction data, and the user behavior data.
The recall module 710 is configured to recall the first question bank according to the target keyword to obtain a second candidate question set.
And the pushing module 712 is configured to determine a first question sequence according to the historical interaction data, the first candidate question set, and the second candidate question set, and push the first question sequence to a display interface corresponding to the user.
In some embodiments, the question pushing device further includes, but is not limited to:
and the target question determining module is used for determining the target question selected by the user in response to the first operation triggered on the display interface.
And the classification matching module is used for performing classification matching on the target problems according to a preset classification problem library to obtain a classification matching result.
And the verification module is used for verifying the user according to the classification matching result or determining a question response corresponding to the target question based on the user attribute information of the user and a preset standard response library.
In some embodiments, the push module comprises:
and the target problem set determining unit is used for obtaining a target problem set according to the first candidate problem set and the second candidate problem set.
And the first question determining unit is used for sequencing the target question set and determining a plurality of first questions from the sequenced target question set.
And the second question determining unit is used for determining a plurality of second questions from the second question bank according to the historical interaction data.
And the combining unit is used for combining the plurality of first questions and the plurality of second questions to obtain a first question sequence.
In some embodiments, the question pushing device further includes, but is not limited to:
and the removing module is used for removing the plurality of first questions from the target question set and removing the plurality of second questions from the second question bank in response to a second operation triggered on the display interface or an operation not triggered in a preset time period by the display interface.
And the problem determining module is used for determining a plurality of third problems from the removed target problem set and determining a plurality of fourth problems from the removed second problem base.
And the combination module is used for combining the plurality of third questions and the plurality of fourth questions to obtain a second question sequence.
And the second pushing module is used for pushing the second question sequence to the display interface.
In some embodiments, the question pushing device further includes, but is not limited to:
and the information acquisition module is used for acquiring operation information triggered by the user on the customer service system and response information generated by the customer service system according to the operation information.
And the user behavior data determining module is used for determining the user behavior data based on the operation information and the response information.
In some embodiments, the question pushing device includes, but is not limited to:
and the interactive information acquisition module is used for acquiring historical interactive information of the user.
And the conversion module is used for converting the historical interaction information into an interaction vector according to a preset pre-training model.
The mining module is used for mining to obtain a problem cluster according to the interaction vector; the problem cluster includes at least one problem.
And the supplement module is used for matching each question in the question cluster according to the first question bank and supplementing the matched question cluster into the first question bank.
In some embodiments, the conversion module comprises:
and the duplicate removal unit is used for removing the duplicate of the historical interaction information to obtain the target interaction information.
And the coding module is used for coding the target interaction information into an interaction vector according to the pre-training model.
The modules in the above problem pushing device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 8. The computer apparatus includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input device. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The input/output interface of the computer device is used for exchanging information between the processor and an external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a problem pushing method. The display unit of the computer device is used for forming a visual picture and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the configuration shown in fig. 8 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the problem pushing method of the following above-mentioned embodiments when executing the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the problem pushing method of the above-described embodiments.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the problem pushing method of the above embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), magnetic Random Access Memory (MRAM), ferroelectric Random Access Memory (FRAM), phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A question pushing method, characterized in that the method comprises:
acquiring user tag data, user behavior data, historical interaction data and service scene characteristics corresponding to the current position of a user;
establishing a baffle rule according to the user label data, the user behavior data and the service scene characteristics;
screening original problems in a preset first problem library according to the baffle rule to obtain a first candidate problem set;
determining a target keyword according to the user tag data, the historical interaction data and the user behavior data;
recalling the first question bank according to the target keywords to obtain a second candidate question set;
and determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set, and pushing the first question sequence to a display interface corresponding to the user.
2. The method of claim 1, further comprising:
determining a target question selected by the user in response to a first operation triggered on a display interface;
carrying out classification matching on the target problems according to a preset classification problem library to obtain a classification matching result;
and verifying the user according to the classification matching result, or determining a question response corresponding to the target question based on the user attribute information of the user and a preset standard response library.
3. The method of claim 1, wherein determining a first question sequence based on the historical interaction data, a first candidate question set, and the second candidate question set comprises:
obtaining a target question set according to the first candidate question set and the second candidate question set;
sequencing the target problem set, and determining a plurality of first problems from the sequenced target problem set;
determining a plurality of second questions from a second question bank according to the historical interaction data;
and obtaining the first question sequence according to the plurality of first questions and the plurality of second questions.
4. The method of claim 3, further comprising:
in response to a second operation triggered on a display interface or an operation not triggered by the display interface in a preset time period, removing the first questions from the target question set and removing the second questions from a second question bank;
determining a plurality of third questions from the removed target question set, and determining a plurality of fourth questions from the removed second question bank;
combining the plurality of third questions and the plurality of fourth questions to obtain a second question sequence;
and pushing the second question sequence to the display interface.
5. The method of claim 1, wherein before the obtaining of the user tag data, the user behavior data, the historical interaction data, and the service scenario features corresponding to the current location of the user, the method further comprises:
acquiring operation information triggered by the user on a customer service system and response information generated by the customer service system according to the operation information;
determining the user behavior data based on the operation information and the response information.
6. The method according to any one of claims 1 to 5, further comprising:
acquiring historical interaction information of a user;
converting the historical interaction information into an interaction vector according to a preset pre-training model;
mining according to the interaction vector to obtain a problem cluster; the problem cluster comprises at least one problem;
and matching each question in the question cluster according to the first question bank, and supplementing the matched question cluster into the first question bank.
7. The method according to claim 6, wherein the converting the historical interaction information into an interaction vector according to a preset pre-training model comprises:
cleaning and repeating the historical interaction information to obtain target interaction information;
and coding the target interaction information into the interaction vector according to the pre-training model.
8. A problem-pushing device, the device comprising:
the data acquisition module is used for acquiring user tag data, user behavior data, historical interaction data of a user and service scene characteristics corresponding to the current position of the user;
the rule establishing module is used for establishing a baffle rule according to the user label data, the user behavior data and the service scene characteristics;
the screening module is used for screening original problems in a preset first problem library according to the baffle rule to obtain a first candidate problem set;
the target keyword determining module is used for determining target keywords according to the user tag data, the historical interaction data and the user behavior data;
the recall module is used for recalling the first question bank according to the target keyword to obtain a second candidate question set;
and the pushing module is used for determining a first question sequence according to the historical interaction data, the first candidate question set and the second candidate question set and pushing the first question sequence to a display interface corresponding to the user.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211499888.9A CN115878780A (en) | 2022-11-28 | 2022-11-28 | Question pushing method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211499888.9A CN115878780A (en) | 2022-11-28 | 2022-11-28 | Question pushing method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115878780A true CN115878780A (en) | 2023-03-31 |
Family
ID=85764269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211499888.9A Pending CN115878780A (en) | 2022-11-28 | 2022-11-28 | Question pushing method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115878780A (en) |
-
2022
- 2022-11-28 CN CN202211499888.9A patent/CN115878780A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Personalized mobile app recommendation: Reconciling app functionality and user privacy preference | |
| US9691035B1 (en) | Real-time updates to item recommendation models based on matrix factorization | |
| CN110909222B (en) | User portrait establishing method and device based on clustering, medium and electronic equipment | |
| CN111553744A (en) | Federal product recommendation method, device, equipment and computer storage medium | |
| US11886556B2 (en) | Systems and methods for providing user validation | |
| CN106030527B (en) | By the system and method for application notification user available for download | |
| CN112258238A (en) | User life value cycle detection method and device and computer equipment | |
| CN111523053A (en) | Information flow processing method and device, computer equipment and storage medium | |
| CN113158047A (en) | Recommendation model training and information pushing method, device, equipment and medium | |
| CN116467525A (en) | Recommendation method, device, equipment and storage medium of business products | |
| CN115758271A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| CN112131502A (en) | Data processing method, data processing apparatus, electronic device, and medium | |
| CN115878780A (en) | Question pushing method and device, computer equipment and storage medium | |
| CN112508075B (en) | DBSCAN clustering method based on transverse federation and related equipment thereof | |
| CN110197056B (en) | Relation network and associated identity recognition method, device, equipment and storage medium | |
| CN116932891A (en) | Resource object display method, device, equipment, storage medium and product | |
| CN116846974A (en) | Service request processing method, device, computer equipment and storage medium | |
| CN115630973A (en) | User data processing method, device, computer equipment and storage medium | |
| CN117407418A (en) | Information acquisition method, information acquisition device, computer apparatus, storage medium, and program product | |
| CN117111800A (en) | Menu configuration method, menu configuration device, computer device, storage medium, and program product | |
| CN116861396A (en) | Login method, device, apparatus, storage medium and program product | |
| CN117667999A (en) | Data pushing method, device, computer equipment and computer readable storage medium | |
| CN116340638A (en) | Method and device for determining interaction result | |
| CN117436972A (en) | Resource object recommendation method, device, computer equipment and storage medium | |
| CN117454215A (en) | Feedback resource configuration method and device and computer equipment |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: 518000 Room 201, building A, No. 1, Qian Wan Road, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong (Shenzhen Qianhai business secretary Co., Ltd.) Applicant after: Zhaolian Consumer Finance Co.,Ltd. Address before: 518000 Room 201, building A, No. 1, Qian Wan Road, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong (Shenzhen Qianhai business secretary Co., Ltd.) Applicant before: MERCHANTS UNION CONSUMER FINANCE Co.,Ltd. Country or region before: China |