CN116150484A - Information pushing method and device, storage medium and electronic equipment - Google Patents
Information pushing method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN116150484A CN116150484A CN202310094301.4A CN202310094301A CN116150484A CN 116150484 A CN116150484 A CN 116150484A CN 202310094301 A CN202310094301 A CN 202310094301A CN 116150484 A CN116150484 A CN 116150484A
- Authority
- CN
- China
- Prior art keywords
- information
- user
- instruction
- determining
- text content
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses an information pushing method, an information pushing device, a storage medium and electronic equipment. The method comprises the following steps: acquiring instruction information input by a user, identifying the instruction information, and obtaining instruction text content; inputting the instruction text content into the interaction model to obtain at least one prediction result; acquiring a group of knowledge bases associated with each user intention information, and respectively inputting initial attribute information sets in at least one user intention information into the associated knowledge bases for searching to obtain at least one group of search results; and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user. According to the method and the device, the problem that in the related art, the accuracy rate of sending feedback information to the user is low due to the fact that the field of instruction information sent by the user cannot be accurately identified is solved.
Description
Technical Field
The present invention relates to the field of information processing, and in particular, to an information pushing method, an information pushing device, a storage medium, and an electronic device.
Background
At present, when a financial institution processes a user service, a dialogue system is generally used to acquire instruction information of the user, identify the instruction information, thereby obtaining feedback information required by the user, and send the feedback information to the user. Dialog systems typically employ domain-intent-slot three-segment structures for user intent partitioning. domain, i.e. domain division, such as weather, alarm clock, etc. The intent is the division of the user intention in a certain field, i.e. the division of the intention in a certain field, such as inquiring the temperature, or inquiring the wind direction, etc. Slot is a Slot for describing details of instruction information of a user. The existing method utilizes the neural network technology to perform multi-classification calculation on instruction information sent by a user. The probability of the instruction information in each field is calculated, then the field with the highest probability is selected, and the subsequent intent-slot division is carried out in the field, so that feedback information corresponding to the instruction information of the user is determined according to the slot.
However, when the currently used information feedback method is used for determining the fields, the same instruction information may hit a plurality of fields, at this time, it cannot be determined from which field to continue to perform the feedback information determination, the field selection is generally performed randomly, and in the case of randomly selecting the field, the information fed back to the user is wrong due to the error of the field selection, so that the user cannot obtain the correct feedback information.
Aiming at the problem that the accuracy rate of the feedback information sent to the user is low because the field of the instruction information sent by the user cannot be accurately identified in the related art, no effective solution is proposed at present.
Disclosure of Invention
The application provides an information pushing method, an information pushing device, a storage medium and electronic equipment, and aims to solve the problem that in the related art, due to the fact that the field of instruction information sent by a user cannot be accurately identified, the accuracy rate of sending feedback information to the user is low.
According to one aspect of the present application, an information pushing method is provided. The method comprises the following steps: acquiring instruction information input by a user, identifying the instruction information, and obtaining instruction text content; inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results; and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user.
Optionally, the interaction model includes a rule component and a model component, and inputting the instruction text content into the interaction model to obtain at least one predicted result includes: inputting the instruction text content into a rule component, and detecting whether the rule component outputs intention information, wherein the rule component comprises a plurality of preset rules and intention information indicated by each preset rule; in the case that the rule component outputs the intention information, determining the intention information output by the rule component as user intention information, and determining a first numerical value as a first score of the instruction text content; in the case that the rule component does not output the intention information, inputting the instruction text content into the model component, obtaining a plurality of intention information and the correlation degree between each intention information and the instruction text content, and determining the correlation degree of each intention information as a first score of the instruction text content.
Optionally, inputting the initial attribute information set in the user intention information into the associated knowledge base for searching, and obtaining at least one group of search results includes: inputting an initial attribute information set in the user intention information into an associated knowledge base for searching, obtaining a search result, and obtaining a preset attribute information set of the search result; determining the coincidence ratio between the initial attribute information set and the preset attribute information set to obtain coincidence ratio information; and determining the search result and the coincidence degree information of the search result as the search result of the knowledge base.
Optionally, determining the degree of coincidence between the initial attribute information set and the preset attribute information set, and obtaining the degree of coincidence information includes: determining the same attribute information in the initial attribute information set and the preset attribute information set to obtain at least one piece of repeated attribute information; determining the attribute type of each piece of repeated attribute information, and determining a first weight of each piece of repeated attribute information according to the attribute type to obtain a plurality of first weights; determining the weight of the non-repeated attribute information in the initial attribute information set as a second weight; calculating a target weight of the initial attribute information set through the first weight and the second weight, and determining the target weight as the coincidence ratio between the initial attribute information set and the preset attribute information set to obtain coincidence ratio information.
Optionally, determining a second score according to each search result and the corresponding first score, respectively, to obtain a plurality of second scores, and determining a target search result with the highest second score from at least one group of search results includes: multiplying the coincidence degree information in each search result with the first score of the corresponding user intention information in sequence to obtain a plurality of second scores; and obtaining a second score with the largest value in the second scores, and determining the search result to which the largest second score belongs as a target search result.
Optionally, obtaining instruction information input by a user, identifying the instruction information, and obtaining instruction text content includes: acquiring the information type of the instruction information, and acquiring an identification program for identifying the text of the information type to obtain a target identification program; identifying instruction information through a target identification program to obtain initial text content; acquiring the text quantity of the initial text content, and judging whether the text quantity of the initial text content is smaller than a preset text quantity or not; inputting the initial text content into an inference model to obtain instruction text content under the condition that the text quantity of the initial text content is smaller than a preset text quantity, wherein the inference model is used for expanding the text quantity of the initial text content; and determining the initial text content as the instruction text content under the condition that the text quantity of the initial text content is larger than or equal to the preset text quantity.
Optionally, after obtaining the knowledge base associated with each user intention information, obtaining at least one group of knowledge bases, and inputting the initial attribute information set in the user intention information into the associated knowledge base for searching, the method further comprises: under the condition that the search result is not obtained, obtaining the user intention information with the maximum first score to obtain target intention information; and acquiring recommendation information associated with the target intention information from the database, and sending the recommendation information to the user.
According to another aspect of the present application, an information pushing apparatus is provided. The device comprises: the first acquisition unit is used for acquiring instruction information input by a user, identifying the instruction information and obtaining instruction text content; the input unit is used for inputting the instruction text content into the interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; the searching unit is used for acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results; and the determining unit is used for respectively determining the second scores according to each search result and the corresponding first score to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user.
According to another aspect of the embodiment of the present invention, there is also provided a computer storage medium for storing a program, where the program controls a device in which the computer storage medium is located to execute an information pushing method when running.
According to another aspect of embodiments of the present invention, there is also provided an electronic device including one or more processors and a memory; the memory stores computer readable instructions, and the processor is configured to execute the computer readable instructions, where the computer readable instructions execute an information push method when executed.
Through the application, the following steps are adopted: acquiring instruction information input by a user, identifying the instruction information, and obtaining instruction text content; inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results; and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user. The problem of the prior art that the accuracy rate of sending feedback information to a user is low because the field of instruction information sent by the user cannot be accurately identified is solved. Through the field of uncertain instruction information, directly after the instruction text content is obtained, determining the intention of a user according to the instruction text content, and determining search results closest to the intention of the user in a plurality of knowledge bases according to the determined intention of the user, so that the search results of each intention in each knowledge base can be obtained, and by calculating the second score of each search result, determining the search result with the highest matching degree with the instruction information of the user in the plurality of search results, obtaining a target search result, and feeding back the target search result to the user, thereby achieving the effect of improving the acquisition accuracy of the search result.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application. In the drawings:
fig. 1 is a flowchart of an information pushing method provided according to an embodiment of the present application;
FIG. 2 is a flow chart of an alternative determination of intent information provided in accordance with an embodiment of the present application;
fig. 3 is a schematic diagram of an information pushing device according to an embodiment of the present application;
fig. 4 is a schematic diagram of an electronic device according to an embodiment of the present application.
Detailed Description
It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the present application described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be noted that, related information (including, but not limited to, user equipment information, user personal information, etc.) and data (including, but not limited to, data for presentation, analyzed data, etc.) related to the present disclosure are information and data authorized by a user or sufficiently authorized by each party. For example, an interface is provided between the system and the relevant user or institution, before acquiring the relevant information, the system needs to send an acquisition request to the user or institution through the interface, and acquire the relevant information after receiving the consent information fed back by the user or institution.
According to an embodiment of the application, an information pushing method is provided.
Fig. 1 is a flowchart of an information pushing method according to an embodiment of the present application. As shown in fig. 1, the method comprises the steps of:
step S101, instruction information input by a user is acquired, the instruction information is identified, and instruction text content is obtained.
Specifically, the instruction information may be instruction information sent when the user obtains information from the target system, for example, when the user inputs an instruction "query a business handling place in a region" in the business system, the instruction information may be input in a form of voice or text input, and at this time, the voice needs to be parsed or the text needs to be identified, so that identification of the instruction information is completed, and then instruction text content in the instruction information is obtained.
Step S102, inputting the instruction text content into the interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set.
Specifically, the interaction model may include one or more engines, each of which may identify the content of the instruction text by using an identification and judgment method preset in the engine, and generate a prediction result according to the identification result, where the prediction result may include user intention information, that is, after identifying the instruction text information, the prediction result corresponds to a plurality of preset user intentions in the engine, so as to determine intention information of the user.
Further, since the interaction model may obtain a plurality of prediction results and the engines corresponding to the prediction results are different, a first score may be determined for each prediction result, and a prediction result with the highest accuracy may be determined from the plurality of prediction results according to the first score.
It should be noted that, the initial attribute information set includes a plurality of initial attribute information, where each initial attribute information is a slot under each intention, that is, instruction details obtained according to instruction text information, for example, when the instruction text information is "query a maximum business handling place in the area a", the information in the slot may include an area of each business place, an area to which the business place belongs, or when the instruction text information is "select alert sound a", the information in the slot may include information such as a name, composer, wind, and year of a. And therefore, the correlation between the selected feedback information and the instruction text information can be determined according to the attribute information.
Step S103, a group of knowledge bases associated with each user intention information is obtained, and initial attribute information sets in at least one user intention information are respectively input into the associated knowledge bases to search, so that at least one group of search results are obtained, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results.
Specifically, each user intention information may be associated with a set of knowledge bases, and each set of knowledge bases may include a plurality of knowledge bases, that is, one user intention information and related information may be stored in a plurality of knowledge bases, for example, in the case that the intention information is "select alert sound is a", if music with a as a name is stored in the a knowledge base, and a recording with a as a name is stored in the B knowledge base, the knowledge base corresponding to the intention information is the a knowledge base and the B knowledge base.
Further, when searching is performed in each knowledge base through the initial attribute information set, a search result is obtained in each knowledge base, and at this time, the initial attribute information set needs to be compared with the search result corresponding to each knowledge base, so as to obtain the coincidence degree of the search result and the initial attribute information, and the search result closest to the indicated text information is determined according to the coincidence degree.
Step S104, determining second scores according to each search result and the corresponding first score respectively, obtaining a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user.
Specifically, under the condition that the coincidence ratio of each search result and the first score of the intention information corresponding to each search result are obtained, the second score of each search result can be determined through the coincidence ratio and the first score, the search result with the highest second score is determined to be the search result closest to the instruction text information, and the search result is sent to the user, so that the effect of accurately feeding back the search information to the user is achieved.
According to the information pushing method, instruction information input by a user is obtained, the instruction information is identified, and instruction text content is obtained; inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results; and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user. The problem of the prior art that the accuracy rate of sending feedback information to a user is low because the field of instruction information sent by the user cannot be accurately identified is solved. Through the field of uncertain instruction information, directly after the instruction text content is obtained, determining the intention of a user according to the instruction text content, and determining search results closest to the intention of the user in a plurality of knowledge bases according to the determined intention of the user, so that the search results of each intention in each knowledge base can be obtained, and by calculating the second score of each search result, determining the search result with the highest matching degree with the instruction information of the user in the plurality of search results, obtaining a target search result, and feeding back the target search result to the user, thereby achieving the effect of improving the acquisition accuracy of the search result.
Optionally, in the information pushing method provided in the embodiment of the present application, the interaction model includes a rule component and a model component, and inputting the instruction text content into the interaction model, the obtaining at least one prediction result includes: inputting the instruction text content into a rule component, and detecting whether the rule component outputs intention information, wherein the rule component comprises a plurality of preset rules and intention information indicated by each preset rule; in the case that the rule component outputs the intention information, determining the intention information output by the rule component as user intention information, and determining a first numerical value as a first score of the instruction text content; in the case that the rule component does not output the intention information, inputting the instruction text content into the model component, obtaining a plurality of intention information and the correlation degree between each intention information and the instruction text content, and determining the correlation degree of each intention information as a first score of the instruction text content.
Specifically, fig. 2 is a flowchart of optional determining intent information provided according to an embodiment of the present application, and as shown in fig. 2, an interaction model may include two components, where each component may separately analyze instruction text content, so as to obtain intent information corresponding to the instruction text content. The interaction model may include a rule component, and a plurality of preset rules may be stored in the rule component, where each preset rule may be used to analyze content of the instruction text, determine whether the instruction text information meets the preset rule, and determine intention information corresponding to the preset rule met by the instruction text information as user intention information. For example, the instruction text content may be "play song a", and at this time, the preset rule a may be: including "play", "song" in the text content, the user intent information may be determined to be the same as the song the user wants to play, rather than playing video.
It should be noted that, in the case where the instruction text content conforms to a certain preset rule, the first score of the intention information is determined to be 1.
Further, in the case that the instruction text content does not conform to any one of the preset rules, determining intention information needs to be performed through a model component, wherein the model component can be a machine learning model, and can determine historical text content with a relevance to the instruction text content being greater than a preset threshold value through analyzing the instruction text content, and determine intention information of the historical text content as user intention information, so that a plurality of user intention information are obtained, and determine the relevance of each historical text content as a first score of each user intention information.
Optionally, in the information pushing method provided in the embodiment of the present application, inputting an initial attribute information set in user intention information into an associated knowledge base for searching, and obtaining at least one set of search results includes: inputting an initial attribute information set in the user intention information into an associated knowledge base for searching, obtaining a search result, and obtaining a preset attribute information set of the search result; determining the coincidence ratio between the initial attribute information set and the preset attribute information set to obtain coincidence ratio information; and determining the search result and the coincidence degree information of the search result as the search result of the knowledge base.
Specifically, after the intention information is obtained, an initial attribute information set of the intention information may be determined according to the indicated text content, for example, when the indicated text content is "play song a", song a may be retrieved in a network or a comprehensive database, so as to obtain an initial attribute information set of song a, where the initial attribute information set may include information such as a singer name, song wind, year, and the like of song a.
Further, after the initial attribute information set is obtained, the initial attribute information set may be input into a plurality of knowledge bases in the target system to perform searching, so that a search result with the highest contact ratio with the initial attribute information set is obtained in each knowledge base, and a plurality of search results are obtained, where each intention information corresponds to a plurality of search results.
Optionally, in the information pushing method provided in the embodiment of the present application, determining a degree of coincidence between an initial attribute information set and a preset attribute information set, where obtaining the degree of coincidence information includes: determining the same attribute information in the initial attribute information set and the preset attribute information set to obtain at least one piece of repeated attribute information; determining the attribute type of each piece of repeated attribute information, and determining a first weight of each piece of repeated attribute information according to the attribute type to obtain a plurality of first weights; determining the weight of the non-repeated attribute information in the initial attribute information set as a second weight; calculating a target weight of the initial attribute information set through the first weight and the second weight, and determining the target weight as the coincidence ratio between the initial attribute information set and the preset attribute information set to obtain coincidence ratio information.
Specifically, when the contact ratio calculation is performed, the weight of each attribute information in the initial attribute information set can be obtained, and the contact ratio of the preset attribute information set obtained in a certain knowledge base and the attribute information in the initial attribute information set, namely, the result after repeated weighted summation of the attribute information, is determined.
For example, the attribute information included in the initial attribute information set includes: A. b, C, D, E, the attribute information included in the preset attribute information set includes: A. b, F, G, the duplicate attribute information is A, B, and when the weight of A, B is determined to be 1 and the weights of c and D, E are determined to be 0, the overlap ratio may be 1+1+0+0+0=2.
Further, an initial value may be determined for each attribute information according to the importance of the attribute information, for example, in the attribute information A, B, C, D, E, the initial value of the attribute information A, C is 1, and the initial value of the attribute information B, D, E is 0.75, and the overlap ratio may be 1×1+0.75×1+1×0+0.75×0+0.75×0=1.75, so as to obtain more accurate overlap ratio information.
Optionally, in the information pushing method provided in the embodiment of the present application, determining, according to each search result and the corresponding first score, a second score, to obtain a plurality of second scores, and determining, from at least one set of search results, a target search result with a highest second score includes: multiplying the coincidence degree information in each search result with the first score of the corresponding user intention information in sequence to obtain a plurality of second scores; and obtaining a second score with the largest value in the second scores, and determining the search result to which the largest second score belongs as a target search result.
Specifically, after the first score and the overlap ratio information of each search result are obtained, the overlap ratio score in each search result can be calculated to be multiplied by the first score of the intention information of the search result, so as to obtain a second score, namely a final score, of the search result, and after the second score of each search result is calculated, the search result with the largest second score is fed back to the user, so that the user can obtain a corresponding feedback result from the instruction text content.
For example, the first score of the search result a is 0.8, the second score is 0.8×0.75=0.6, the first score of the search result B is 0.7, the overlap ratio information is 0.9, the second score is 0.9×0.7=0.63, and the search result B is taken as the target search result of the feedback user.
Optionally, in the information pushing method provided in the embodiment of the present application, obtaining instruction information input by a user, and identifying the instruction information, where obtaining instruction text content includes: acquiring the information type of the instruction information, and acquiring an identification program for identifying the text of the information type to obtain a target identification program; identifying instruction information through a target identification program to obtain initial text content; acquiring the text quantity of the initial text content, and judging whether the text quantity of the initial text content is smaller than a preset text quantity or not; inputting the initial text content into an inference model to obtain instruction text content under the condition that the text quantity of the initial text content is smaller than a preset text quantity, wherein the inference model is used for expanding the text quantity of the initial text content; and determining the initial text content as the instruction text content under the condition that the text quantity of the initial text content is larger than or equal to the preset text quantity.
Specifically, since the user can input the instruction through various instruction input modes, such as voice data, text input, gesture input, and the like, the target system needs to acquire the information type of the instruction information first, and determine the corresponding recognition program according to the information type, so that the instruction information can be converted into unified instruction text content according to the recognition program, and the instruction text content is analyzed to obtain an analysis result.
It should be noted that after the instruction text content is obtained through the instruction information, the instruction text content is possibly less due to the fact that the instruction information is too less, and further the obtained instruction text content is less, so that the intention information is wrong according to the instruction text content, and the determination of the feedback information is affected.
Optionally, in the information pushing method provided in the embodiment of the present application, after obtaining a knowledge base associated with each user intention information, obtaining at least one set of knowledge bases, and inputting an initial attribute information set in the user intention information into the associated knowledge base for searching, the method further includes: under the condition that the search result is not obtained, obtaining the user intention information with the maximum first score to obtain target intention information; and acquiring recommendation information associated with the target intention information from the database, and sending the recommendation information to the user.
It should be noted that, after the intention information is obtained, if the search result is obtained in the multiple knowledge bases, it is characterized that feedback information corresponding to the instruction information does not exist in the search result, and at this time, there may be an error in the instruction information input by the user, so that the intention information with the largest first score may be obtained, and the recommendation information corresponding to the intention information may be sent to the user, thereby enabling the user to perform content selection by himself.
For example, when it is recognized that the intention information of the user is played music a, but a preset attribute information set corresponding to an initial attribute information set related to played music a does not exist in the knowledge base, at this time, a music list stored in the database is directly fed back to the user, so that the user can select music by himself.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
The embodiment of the application also provides an information pushing device, and the information pushing device of the embodiment of the application can be used for executing the information pushing method provided by the embodiment of the application. The following describes an information pushing device provided in the embodiment of the present application.
Fig. 3 is a schematic diagram of an information pushing device according to an embodiment of the present application. As shown in fig. 3, the apparatus includes: a first acquisition unit 31, an input unit 32, a search unit 33, a determination unit 34.
The first obtaining unit 31 is configured to obtain instruction information input by a user, and identify the instruction information, so as to obtain instruction text content.
An input unit 32, configured to input the instruction text content into the interaction model, and obtain at least one prediction result, where the prediction result includes user intention information, and a first score of the user intention information, and the user intention information includes an initial attribute information set.
The searching unit 33 is configured to obtain a set of knowledge bases associated with each user intention information, and input an initial set of attribute information in at least one user intention information into the associated knowledge bases respectively for searching, so as to obtain at least one set of search results, where each set of knowledge bases includes a plurality of knowledge bases, and each set of search results includes a plurality of search results.
And the determining unit 34 is configured to determine second scores according to each search result and the corresponding first score, obtain a plurality of second scores, determine a target search result with the highest second score from at least one group of search results, and send the target search result to the user.
According to the information pushing device provided by the embodiment of the application, the first obtaining unit 31 is used for obtaining the instruction information input by the user and identifying the instruction information to obtain the instruction text content; the input unit 32 inputs the instruction text content into the interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; the searching unit 33 obtains a set of knowledge bases associated with each user intention information, and inputs an initial attribute information set in at least one user intention information into the associated knowledge bases respectively for searching to obtain at least one set of search results, wherein each set of knowledge bases comprises a plurality of knowledge bases, and each set of search results comprises a plurality of search results; the determining unit 34 determines second scores according to each search result and the corresponding first score, obtains a plurality of second scores, determines a target search result with the highest second score from at least one group of search results, and sends the target search result to the user. The problem of the prior art that the accuracy rate of sending feedback information to a user is low because the field of instruction information sent by the user cannot be accurately identified is solved. Through the field of uncertain instruction information, directly after the instruction text content is obtained, determining the intention of a user according to the instruction text content, and determining search results closest to the intention of the user in a plurality of knowledge bases according to the determined intention of the user, so that the search results of each intention in each knowledge base can be obtained, and by calculating the second score of each search result, determining the search result with the highest matching degree with the instruction information of the user in the plurality of search results, obtaining a target search result, and feeding back the target search result to the user, thereby achieving the effect of improving the acquisition accuracy of the search result.
Optionally, in the information pushing apparatus provided in the embodiment of the present application, the interaction model includes a rule component and a model component, and the input unit 32 includes: the detection module is used for inputting the instruction text content into the rule assembly and detecting whether the rule assembly outputs intention information, wherein the rule assembly comprises a plurality of preset rules and intention information indicated by each preset rule; a first determining module for determining the intention information output by the rule component as user intention information and determining a first value as a first score of the instruction text content in the case that the rule component outputs the intention information; and the second determining module is used for inputting the instruction text content into the model component under the condition that the rule component does not output the intention information, obtaining a plurality of intention information and the correlation degree between each intention information and the instruction text content, and determining the correlation degree of each intention information as a first score of the instruction text content.
Optionally, in the information pushing device provided in the embodiment of the present application, the search unit 33 includes: the searching module is used for inputting the initial attribute information set in the user intention information into the associated knowledge base for searching, obtaining a searching result and obtaining a preset attribute information set of the searching result; the third determining module is used for determining the coincidence degree between the initial attribute information set and the preset attribute information set to obtain coincidence degree information; and the fourth determining module is used for determining the search result and the coincidence degree information of the search result as the search result of the knowledge base.
Optionally, in the information pushing apparatus provided in the embodiment of the present application, the third determining module includes: the first determining submodule is used for determining the same attribute information in the initial attribute information set and the preset attribute information set to obtain at least one piece of repeated attribute information; the second determining submodule is used for determining the attribute type of each piece of repeated attribute information, determining the first weight of each piece of repeated attribute information according to the attribute type and obtaining a plurality of first weights; a third determining sub-module, configured to determine a weight of non-duplicate attribute information in the initial attribute information set as a second weight; the calculating sub-module is used for calculating the target weight of the initial attribute information set through the first weight and the second weight, determining the target weight as the contact ratio between the initial attribute information set and the preset attribute information set, and obtaining contact ratio information.
Optionally, in the information pushing device provided in the embodiment of the present application, the determining unit 34 includes: the computing module is used for multiplying the coincidence degree information in each search result with the first score of the corresponding user intention information in sequence to obtain a plurality of second scores; the first acquisition module is used for acquiring a second score with the largest value in the second scores and determining a search result to which the largest second score belongs as a target search result.
Optionally, in the information pushing apparatus provided in the embodiment of the present application, the first obtaining unit 31 includes: the second acquisition module is used for acquiring the information type of the instruction information and acquiring an identification program for identifying the text of the information type to obtain a target identification program; the identification module is used for identifying instruction information through a target identification program to obtain initial text content; the judging module is used for acquiring the text quantity of the initial text content and judging whether the text quantity of the initial text content is smaller than a preset text quantity or not; the input module is used for inputting the initial text content into the inference model to obtain instruction text content under the condition that the text quantity of the initial text content is smaller than the preset text quantity, wherein the inference model is used for expanding the text quantity of the initial text content; and a fifth determining module, configured to determine the initial text content as the instruction text content when the text amount of the initial text content is greater than or equal to the preset text amount.
Optionally, in the information pushing device provided in the embodiment of the present application, the device further includes: the second acquisition unit is used for acquiring the user intention information with the maximum first score under the condition that the search result is not obtained, so as to obtain target intention information; and the recommending unit is used for acquiring the recommending information associated with the target intention information from the database and sending the recommending information to the user.
The information pushing device includes a processor and a memory, the first obtaining unit 31, the input unit 32, the searching unit 33, the determining unit 34, and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one kernel, and the problem that the accuracy rate of sending feedback information to a user is low due to the fact that the field of instruction information sent by the user cannot be accurately identified in the related art is solved by adjusting kernel parameters.
The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
The embodiment of the invention provides a computer readable storage medium, on which a program is stored, which when executed by a processor, implements the information push method.
The embodiment of the invention provides a processor which is used for running a program, wherein the information pushing method is executed when the program runs.
As shown in fig. 4, an embodiment of the present invention provides an electronic device, where the electronic device 40 includes a processor, a memory, and a program stored on the memory and executable on the processor, and when the processor executes the program, the following steps are implemented: acquiring instruction information input by a user, identifying the instruction information, and obtaining instruction text content; inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results; and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user. The device herein may be a server, PC, PAD, cell phone, etc.
The present application also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with the method steps of: acquiring instruction information input by a user, identifying the instruction information, and obtaining instruction text content; inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set; acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results; and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from at least one group of search results, and sending the target search result to the user.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.
Claims (10)
1. An information pushing method is characterized by comprising the following steps:
acquiring instruction information input by a user, and identifying the instruction information to obtain instruction text content;
inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set;
Acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results;
and respectively determining second scores according to each search result and the corresponding first scores to obtain a plurality of second scores, determining a target search result with the highest second score from the at least one group of search results, and sending the target search result to the user.
2. The method of claim 1, wherein the interaction model includes a rules component and a model component, wherein inputting the instructional text content into the interaction model results in at least one predicted outcome includes:
inputting the instruction text content into the rule component, and detecting whether the rule component outputs intention information, wherein the rule component comprises a plurality of preset rules and intention information indicated by each preset rule;
determining the intention information output by the rule component as the user intention information and determining a first value as a first score of the instruction text content in the case that the rule component outputs the intention information;
And under the condition that the rule component does not output the intention information, inputting the instruction text content into the model component, obtaining a plurality of intention information and the relevance between each intention information and the instruction text content, and determining the relevance of each intention information as a first score of the instruction text content.
3. The method of claim 1, wherein the step of inputting the initial set of attribute information in the at least one user intention information into the associated knowledge base for searching, respectively, to obtain at least one set of search results comprises:
inputting an initial attribute information set in the user intention information into an associated knowledge base for searching, obtaining a search result, and obtaining a preset attribute information set of the search result;
determining the coincidence degree between the initial attribute information set and the preset attribute information set to obtain coincidence degree information;
and determining the search result and the coincidence degree information of the search result as search results of a knowledge base.
4. A method according to claim 3, wherein determining the degree of overlap between the initial set of attribute information and the set of preset attribute information, the degree of overlap information comprising:
Determining the same attribute information in the initial attribute information set and the preset attribute information set to obtain at least one piece of repeated attribute information;
determining the attribute type of each piece of repeated attribute information, and determining a first weight of each piece of repeated attribute information according to the attribute type to obtain a plurality of first weights;
determining the weight of the non-repeated attribute information in the initial attribute information set as a second weight;
calculating a target weight of the initial attribute information set through the first weight and the second weight, and determining the target weight as the coincidence degree between the initial attribute information set and the preset attribute information set to obtain coincidence degree information.
5. A method according to claim 3, wherein determining a second score from each search result and the corresponding first score, respectively, to obtain a plurality of second scores, and determining a target search result having a highest second score from the at least one set of search results comprises:
multiplying the coincidence degree information in each search result with the first score of the corresponding user intention information in sequence to obtain a plurality of second scores;
and obtaining a second score with the largest value in the second scores, and determining a search result to which the largest second score belongs as the target search result.
6. The method of claim 1, wherein obtaining instruction information entered by a user and identifying the instruction information, the obtaining instruction text content comprising:
acquiring the information type of the instruction information, and acquiring an identification program for identifying the text of the information type to obtain a target identification program;
identifying the instruction information through the target identification program to obtain initial text content;
acquiring the text quantity of the initial text content, and judging whether the text quantity of the initial text content is smaller than a preset text quantity or not;
inputting the initial text content into an inference model to obtain the instruction text content under the condition that the text quantity of the initial text content is smaller than the preset text quantity, wherein the inference model is used for expanding the text quantity of the initial text content;
and determining the initial text content as the instruction text content under the condition that the text quantity of the initial text content is larger than or equal to the preset text quantity.
7. The method of claim 1, wherein after obtaining a set of knowledge bases associated with each user intention information and separately inputting the initial set of attribute information in the at least one user intention information into the associated knowledge bases for searching, the method further comprises:
Under the condition that the search result is not obtained, obtaining the user intention information with the maximum first score to obtain target intention information;
and acquiring recommendation information associated with the target intention information from a database, and sending the recommendation information to the user.
8. An information pushing apparatus, characterized by comprising:
the first acquisition unit is used for acquiring instruction information input by a user, identifying the instruction information and obtaining instruction text content;
the input unit is used for inputting the instruction text content into an interaction model to obtain at least one prediction result, wherein the prediction result comprises user intention information and a first score of the user intention information, and the user intention information comprises an initial attribute information set;
the searching unit is used for acquiring a group of knowledge bases associated with each user intention information, and respectively inputting an initial attribute information set in at least one user intention information into the associated knowledge bases to search to obtain at least one group of search results, wherein each group of knowledge bases comprises a plurality of knowledge bases, and each group of search results comprises a plurality of search results;
and the determining unit is used for respectively determining second scores according to each search result and the corresponding first score to obtain a plurality of second scores, determining a target search result with the highest second score from the at least one group of search results, and sending the target search result to the user.
9. A computer storage medium for storing a program, wherein the program when run controls a device in which the computer storage medium is located to perform the information pushing method according to any one of claims 1 to 7.
10. An electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the information push method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310094301.4A CN116150484A (en) | 2023-02-08 | 2023-02-08 | Information pushing method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310094301.4A CN116150484A (en) | 2023-02-08 | 2023-02-08 | Information pushing method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116150484A true CN116150484A (en) | 2023-05-23 |
Family
ID=86355719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310094301.4A Pending CN116150484A (en) | 2023-02-08 | 2023-02-08 | Information pushing method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116150484A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117556061A (en) * | 2023-11-20 | 2024-02-13 | 曾昭涵 | Text output method and device, electronic equipment and storage medium |
-
2023
- 2023-02-08 CN CN202310094301.4A patent/CN116150484A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117556061A (en) * | 2023-11-20 | 2024-02-13 | 曾昭涵 | Text output method and device, electronic equipment and storage medium |
| CN117556061B (en) * | 2023-11-20 | 2024-05-24 | 曾昭涵 | Text output method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200374589A1 (en) | User feature generation method and apparatus, device, and computer-readable storage medium | |
| US9582757B1 (en) | Scalable curation system | |
| US20170199936A1 (en) | Methods and systems for search engines selection & optimization | |
| CN110968663B (en) | Answer display method and device of question-answering system | |
| US20180052824A1 (en) | Task identification and completion based on natural language query | |
| US11386095B2 (en) | Natural language querying of data in a structured context | |
| CN111309965B (en) | Audio matching method, device, computer equipment and storage medium | |
| US20170054779A1 (en) | Media Feature Determination for Internet-based Media Streaming | |
| US11354754B2 (en) | Generating self-support metrics based on paralinguistic information | |
| CN110163376B (en) | Sample detection method, media object identification method, device, terminal and medium | |
| US11163812B2 (en) | Classification apparatus and classification method | |
| CN109408799B (en) | Semantic decision method and system | |
| CN114338413B (en) | Method, device and storage medium for determining topological relation of equipment in network | |
| CN116150484A (en) | Information pushing method and device, storage medium and electronic equipment | |
| CN111081225B (en) | Skill voice awakening method and device | |
| CN116150306A (en) | Training method of question-answering robot, question-answering method and device | |
| CN116028626A (en) | Text matching method and device, storage medium and electronic equipment | |
| CN117520645A (en) | User determination method and device based on financial products and electronic equipment | |
| EP4217947A1 (en) | Counterfeit item detection system | |
| CN113010788A (en) | Information pushing method and device, electronic equipment and computer readable storage medium | |
| CN115599891B (en) | Method, device and equipment for determining abnormal dialogue data and readable storage medium | |
| CN113593543B (en) | Intelligent loudspeaker voice service system, method, device and equipment | |
| CN117421418A (en) | Text searching method and device based on keywords and electronic equipment | |
| CN117633282A (en) | Query method and device for financial products, storage medium and electronic equipment | |
| CN117828031A (en) | Construction method and device of dialogue scene, storage medium and electronic 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 |