CN113468420B - Product recommendation method and system - Google Patents
Product recommendation method and system Download PDFInfo
- Publication number
- CN113468420B CN113468420B CN202110726350.6A CN202110726350A CN113468420B CN 113468420 B CN113468420 B CN 113468420B CN 202110726350 A CN202110726350 A CN 202110726350A CN 113468420 B CN113468420 B CN 113468420B
- Authority
- CN
- China
- Prior art keywords
- dialogue
- product
- clause
- content
- classification model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000013145 classification model Methods 0.000 claims abstract description 47
- 230000003068 static effect Effects 0.000 claims abstract description 27
- 230000008569 process Effects 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims 2
- 238000005516 engineering process Methods 0.000 abstract description 6
- 230000008901 benefit Effects 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000009193 crawling Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- RWSOTUBLDIXVET-UHFFFAOYSA-N Dihydrogen sulfide Chemical compound S RWSOTUBLDIXVET-UHFFFAOYSA-N 0.000 description 1
- 102100038367 Gremlin-1 Human genes 0.000 description 1
- 101001032872 Homo sapiens Gremlin-1 Proteins 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—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
- 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
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The application relates to a method and a system for recommending products, wherein the method for recommending the products comprises the following steps: acquiring dialogue content, extracting dialogue characteristics in the dialogue content, acquiring client static characteristics in an intelligent dialogue system, and splicing the dialogue characteristics and the client static characteristics; inputting the spliced result into a classification model, and recalling a plurality of first candidate products by the classification model, wherein the classification model is trained according to dialogue characteristics and product selection results in historical dialogue contents and according to static characteristics of clients when the classification model is trained; according to the method and the device for recommending the products, the problem that the success rate of product recommendation is low when the intelligent dialogue system recommends the products to the clients in the related technology is solved, and the marketing efficiency of the company is improved.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a method and a system for recommending products.
Background
With the rapid development of artificial intelligence technology, intelligent dialogue systems have developed, and most intelligent dialogue systems currently function to answer a customer question, for example, in an intelligent marketing type intelligent dialogue system, in the case that a customer consults a specified product with an intelligent system, the intelligent dialogue system recalls similar consultation sentences according to the degree of similarity according to the consultation contents of the customer, determines reply contents corresponding to the consultation sentences, and feeds back the reply contents to the customer.
Besides a simple question-answering function, in the related technology, the intelligent dialogue system can actively recommend products to clients, but because the client portrait system content of the intelligent dialogue system is relatively thin, the client static characteristics are only surname, gender and mobile phone number of the clients in many times, in the related technology, the intelligent dialogue system presets target products to be recommended and recommending processes, and recommends the same target products to different clients, but the matching degree of the products recommended in the mode and the clients is low, and the probability of successful product recommendation is low.
Aiming at the problem that the success rate of product recommendation is low when an intelligent dialogue system recommends products to clients in the related art, no effective solution is proposed yet.
Disclosure of Invention
The embodiment of the application provides a method and a system for recommending products, which are used for at least solving the problem that the success rate of product recommendation is lower when an intelligent dialogue system recommends products to clients in the related technology.
In a first aspect, an embodiment of the present application provides a method for recommending a product, applied to an intelligent dialogue system, where the method includes:
acquiring dialogue content, extracting dialogue characteristics in the dialogue content, acquiring client static characteristics in the intelligent dialogue system, and splicing the dialogue characteristics and the client static characteristics;
inputting a splicing result into a classification model, wherein the classification model recalls a plurality of first candidate products, and training the classification model according to dialogue characteristics and product selection results in historical dialogue contents and according to the static characteristics of the clients when training the classification model;
and sorting the plurality of first candidate products according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
In some of these embodiments, the dialog features include intent classification results, and the process of extracting dialog features in the dialog content includes:
splitting the dialogue content into a plurality of clauses, determining whether the clauses exist in a question-format list, and if so, determining an intention classification result of the clauses according to the question-format list;
if not, inputting the clause into a classification model, and outputting an intention classification result of the clause by the classification model.
In some of these embodiments, the intent classification result includes a product type and product attributes of the customer consultation, and after the extracting of the dialog features in the dialog content, the method includes:
determining product names corresponding to the dialogue features according to the dialogue features and the association relation of prestored product types, product names and product attributes to obtain a plurality of second candidate products;
summarizing the first candidate product and the second candidate product to obtain a total set of candidate products;
and sorting the plurality of candidate product total sets according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
In some of these embodiments, the method comprises:
after the dialogue content is acquired, dialogue content information is extracted in a dialogue round, and the information is filled into a preset slot of a global dialogue information table;
and before determining whether the clause exists in the list of question patterns, determining whether the clause structure is complete, and if not, complementing the clause according to the information of the history turns in the global information table.
In some of these embodiments, after the splitting the dialog content into a plurality of clauses, before the determining whether the clause is present in the list of question patterns, the method includes: extracting the entity in the clause, determining the entity type of the entity, and replacing the entity in the clause with the entity type.
In some of these embodiments, after the extracting the entity in the clause, the method further comprises: and determining whether the description of the entity meets a preset standard, and if not, changing the description of the entity into a standard description in a sentence.
In a second aspect, an embodiment of the present application provides a system for recommending products, applied to an intelligent dialogue system, where the system includes:
the acquisition module is used for acquiring dialogue content, extracting dialogue characteristics in the dialogue content, acquiring customer static characteristics in the intelligent dialogue system and splicing the dialogue characteristics and the customer static characteristics;
a recall module, configured to input a splice result into a classification model, where the classification model recalls a plurality of first candidate products, where the classification model is trained according to dialogue features and product selection results in historical dialogue content and according to the customer static features when the classification model is trained;
the determining module is used for sorting the plurality of first candidate products according to a preset product recommending strategy and determining recommended products according to the product sorting.
In some embodiments, the dialog features include intent classification results, and in the obtaining module, the process of extracting dialog features in the dialog content includes:
splitting the dialogue content into a plurality of clauses, determining whether the clauses exist in a question-format list, and if so, determining an intention classification result of the clauses according to the question-format list;
if not, inputting the clause into a classification model, and outputting an intention classification result of the clause by the classification model.
In some of these embodiments, the intent classification result includes a product type and product attributes of the customer consultation;
the recall module is further configured to determine, after the extracting the dialogue feature in the dialogue content, a product name corresponding to the dialogue feature according to the dialogue feature and according to a pre-stored association relationship between a product type, a product name and a product attribute, so as to obtain a plurality of second candidate products;
the determining module is further used for summarizing the first candidate product and the second candidate product to obtain a candidate product total set; and sorting the plurality of candidate product total sets according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
In some of these embodiments, the acquisition module is further configured to:
after the dialogue content is acquired, dialogue content information is extracted in a dialogue round, and the information is filled into a preset slot of a global dialogue information table;
and before determining whether the clause exists in the list of question patterns, determining whether the clause structure is complete, and if not, complementing the clause according to the information of the history turns in the global information table.
Compared with the related art, the method for recommending the product provided by the embodiment of the application extracts the dialogue characteristics in the dialogue content by acquiring the dialogue content, and acquires the client static characteristics in the intelligent dialogue system, and splices the dialogue characteristics and the client static characteristics; inputting the spliced result into a classification model, and recalling a plurality of first candidate products by the classification model, wherein the classification model is trained according to dialogue characteristics and product selection results in historical dialogue contents and according to static characteristics of clients when the classification model is trained; according to a preset product recommendation strategy, a plurality of first candidate products are ranked, recommended products are determined according to the product ranking, the problem that the success rate of product recommendation is low when an intelligent dialogue system recommends products to clients in the related technology is solved, and marketing efficiency of a company is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a schematic view of an application environment of a method of product recommendation according to an embodiment of the present application;
FIG. 2 is a flow chart of a method of product recommendation according to a first embodiment of the present application;
FIG. 3 is a flow chart of extracting dialogue features in dialogue content according to a second embodiment of the application;
FIG. 4 is a flow chart of a method of product recommendation according to a third embodiment of the present application;
FIG. 5 is a flow chart of a method of product recommendation according to a fourth embodiment of the present application;
FIG. 6 is a flow chart of a method of product recommendation according to a fifth embodiment of the present application;
fig. 7 is a block diagram of a system for product recommendation according to a sixth embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described and illustrated below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden on the person of ordinary skill in the art based on the embodiments provided herein, are intended to be within the scope of the present application.
It is apparent that the drawings in the following description are only some examples or embodiments of the present application, and it is possible for those of ordinary skill in the art to apply the present application to other similar situations according to these drawings without inventive effort. Moreover, it should be appreciated that while such a development effort might be complex and lengthy, it would nevertheless be a routine undertaking of design, fabrication, or manufacture for those of ordinary skill having the benefit of this disclosure, and thus should not be construed as having the benefit of this disclosure.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly and implicitly understood by those of ordinary skill in the art that the embodiments described herein can be combined with other embodiments without conflict.
Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar terms herein do not denote a limitation of quantity, but rather denote the singular or plural. The terms "comprising," "including," "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to only those steps or elements but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. The terms "connected," "coupled," and the like in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as used herein refers to two or more. "and/or" describes an association relationship of an association object, meaning that there may be three relationships, e.g., "a and/or B" may mean: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship. The terms "first," "second," "third," and the like, as used herein, are merely distinguishing between similar objects and not representing a particular ordering of objects.
The method for recommending the product provided by the application can be applied to an application environment shown in fig. 1, fig. 1 is an application environment schematic diagram of the method for recommending the product according to the embodiment of the application, as shown in fig. 1, a terminal 101 is deployed with an intelligent dialogue system, a server 102 obtains dialogue content of the terminal 101 through a network and operates the method for recommending the product to obtain a product recommendation result, the intelligent dialogue system obtains the product recommendation result and recommends the product to a client, the terminal 101 can be but not limited to various dialogue robots, personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server 102 can be realized by an independent server or a server cluster formed by a plurality of servers.
The present embodiment provides a method for recommending products, which is applied to an intelligent dialogue system, and fig. 2 is a flowchart of a method for recommending products according to a first embodiment of the present application, as shown in fig. 2, where the flowchart includes the following steps:
step S201, dialogue content is obtained, dialogue characteristics in the dialogue content are extracted, customer static characteristics in an intelligent dialogue system are obtained, dialogue characteristics and customer static characteristics are spliced, wherein after the dialogue content is obtained, correction can be carried out on dialogue texts, then dialogue texts are input into a pre-trained FastText word vector model, vectors corresponding to each token are returned, and the vectors are used as token email of a subsequent model;
step S202, inputting a splicing result into a classification model, and recalling a plurality of first candidate products by the classification model, wherein when the classification model is trained, the classification model is trained according to dialogue characteristics and product selection results in historical dialogue contents and according to static characteristics of clients;
step S203, sorting the plurality of first candidate products according to a preset product recommendation policy, and determining recommended products according to the product sorting, where the recommendation policy relates to product attributes, customer attributes, marketing policies, operation policies, and the like, and for example, the recommendation policy is set as follows: and recommending the product with the marketing strategy most preferentially, and ranking the product A first according to the recommending strategy under the condition that the product A has a unified storage activity in the current period.
Through steps S201 to S203, compared with the problem that in the related art, when the intelligent dialogue system recommends a product to a client, the success rate of product recommendation is lower, this embodiment fully uses the opportunity of directly reaching the client, new relevant features of the client are discovered from dialogue contents of the client and the intelligent dialogue system which are dynamically changed, the new relevant features of the client are combined with client static features, so that the matching degree of the recommended product obtained in this embodiment and the client is higher, secondly, as the client is in continuous change, compared with the client static features, the latest situation of the client is reflected in the dialogue process, the instantaneity is higher, so that the product recommended to the client is closer to the current situation of the client, and thirdly, this embodiment trains a classification model according to dialogue features and product selection results in the history dialogue contents, fully uses history data which have been recommended to other clients and successfully selected by other clients, on the basis of the history data, the possibility that the recommended product is adopted by the client is higher, on the basis of the history data, on the embodiment improves the success rate of recommending the relevant products of the intelligent dialogue system, and the success rate of the intelligent dialogue system is lower than the product recommended to the client.
In some of these embodiments, the dialog features include intent classification results, and fig. 3 is a flowchart of extracting dialog features in dialog content according to a second embodiment of the present application, as shown in fig. 3, the flowchart including the steps of:
step S301, splitting dialogue content into a plurality of clauses, and determining whether the clauses exist in a list of question patterns, if so, determining an intention classification result of the clauses according to the list of question patterns, wherein the intention classification result can be in a hierarchical structure, for example, a consultation class (a first layer) comprises consultation product information (a second layer) and the consultation product information (a second layer) comprises consultation interest (a third layer);
in step S302, if not, the clause is input into a classification model, and the classification model outputs the intention classification result of the clause, wherein the classification model may be an LSTM and Attention combined architecture, so as to fully take the key information and the context order information into consideration.
Through steps S301 to S302, the question patterns with different intentions are combed in advance in the embodiment to realize accurate matching of dialogue contents, so that the real intentions behind the language of the customer can be deeply mined in the dialogue process with the customer, and the matching degree of the product recommendation result obtained in the embodiment and the customer is higher.
In some of these embodiments, after splitting the dialog content into multiple clauses, it is determined whether the clause is present in the list of question patterns prior to: the entity in the clause can be extracted, the entity type of the entity is determined, and the entity in the clause is replaced by the entity type; by replacing the entities in the clauses with entity types, the generalization capability of the model is improved, so that the model focuses more on the relationship among the entity types rather than on a single entity.
Further, after extracting the entity in the sentence, it is further required to determine whether the description of the entity meets the preset standard, if not, the description of the entity is changed into the standard description in the sentence; for example, "build line", "build in middle" is normalized to "chinese build bank", and "tomorrow" is normalized to the specific date of the next day, etc., and for example, when the dialogue content of the client is acquired as "can go to and transact on the next week? Under the condition of' the intelligent dialogue system normalizes the next week into a specific time period, searches the pre-stored product knowledge graph database for the handling expiration date of the product, and determines whether the time period is before the handling expiration date; through carrying out normalization processing on the identified entity, the meaning of the sentence can be understood more accurately, the situation that the user cannot answer the entity replacement expression is avoided, smooth communication between the user and the intelligent dialogue system is ensured, and meanwhile, when product recommendation is carried out, the probability of finding matched data in prestored data is improved, so that a proper product is recommended for the user.
Since the intent itself also implies a customer's propensity to select products, in some embodiments where the intent classification result includes the type of product and product attributes of the customer's consultation, these product-related intents may be utilized to make candidate product recalls, FIG. 4 is a flow chart of a method of product recommendation according to a third embodiment of the present application, as shown in FIG. 4, after extracting dialog features in the dialog content, the flow includes the steps of:
step S401, determining a product name corresponding to the dialogue feature according to the dialogue feature and according to the pre-stored association relationship between the product type, the product name and the product attribute, to obtain a plurality of second candidate products, where optionally, the association relationship between the product type, the product name and the product attribute may be pre-stored in a knowledge graph, and the construction process of the knowledge graph includes: firstly, carrying out knowledge graph data structure design, namely constructing a graph based on labels and attributes by adopting a Neo4j data model, for example, dividing entities into different types of deposit, financing, loan, credit card and the like, wherein the different types of entities correspond to different attributes, and deposit products mainly comprise attributes such as product names, organization names, applicable groups, higher line standard rate rising, deposit amount, highest interest rate, highest deposit period, on-line or off-line or the like; secondly, crawling knowledge spectrum data sources through channels such as official pages of business parties, and manually confirming data which can be put in storage after preliminary arrangement; thirdly, executing data storage, wherein the data storage can use an open source database Hugegraph, and the bottom layer is based on an Apache TinkerPop3 frame to support a distributed type; after the knowledge graph database is built, the bottom layer can adopt Gremlin language to inquire by using Java Client, and the product names corresponding to the dialogue features are determined according to the association relation of the product types, the product names and the product attributes in the knowledge graph to obtain a plurality of second candidate products;
step S402, summarizing the first candidate product and the second candidate product to obtain a total set of candidate products;
step S403, sorting the total set of the plurality of candidate products according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
Through steps S401 to S403, in this embodiment, considering that there is a situation that there is less or no history dialogue content related to a part of products when training the classification model, in addition to obtaining a first candidate product according to the dialogue feature and the selection result of the history dialogue content, a second candidate product conforming to the dialogue feature is obtained according to the obtained dialogue feature and the data in the pre-stored product knowledge graph, and the first candidate product and the second candidate product are summarized to obtain a candidate product total set, so that candidate products are enriched, and when a plurality of candidate product total sets are ranked according to a preset product recommendation policy, a more suitable recommended product can be determined.
Considering that a complete representation of a customer's intent sometimes involves multiple rounds of conversations, many of which are based on previous conversational context records, resulting in partial information being omitted from the customer, sometimes by pronouns, sometimes even pronouns, e.g., a robot introducing a loan product, a customer asking "how much interest rate", a robot answering, a customer saying "this interest rate is somewhat high", a real intent is to want a product that is less interest than the product, and to let the intelligent conversational system actually know the real intent of the customer, the first step to do is information completion, in some of which fig. 5 is a flowchart of a method of product recommendation according to a fourth embodiment of the present application, as shown in fig. 5, the flowchart comprising the steps of:
step S501, after obtaining dialogue content, extracting dialogue content information in a dialogue round, and filling the information into a preset slot of a global dialogue information table;
step S502, before determining whether the clause exists in the list of question patterns, determining whether the clause structure is complete, if not, completing the clause according to the information of the history turns in the global information table.
Through steps 501 to 502, the client dialogue of each round is filled in the global dialogue information table, on one hand, the global intention change of the client can be conveniently tracked, the condition that the sentence is ambiguous when the sentence is identified due to the omission of the information of the client is avoided, on the other hand, the latest intention of the client can be conveniently known, alternatively, a product is taken as an entry point in the whole dialogue process, the dialogue is continuously conducted around the intention before the intention of the client on the global information table is transferred, new dialogue content is continuously acquired, and the content information is updated to the global dialogue information table, so that some products which are more suitable for the client can be continuously recommended to the client according to the global information table.
In some embodiments, fig. 6 is a flowchart of a method for recommending products according to a fifth embodiment of the present application, as shown in fig. 6, in each round of dialogue, dialogue speech is acquired, text is obtained after ASR, the dialogue text is understood, the understanding process includes text error correction, embellishing extraction, named entity recognition and intention recognition, relevant product recall is performed in a knowledge graph database of the products according to intention classification features obtained by intention recognition, the recall process includes architecture (Schema) design, data crawling, data storage and data query, finally, the recalled products are ordered, recommended products are obtained after aspects of product attributes, customer attributes, marketing strategies, operation strategies and the like are integrated, and the dialogue information of this round is updated in a global dialogue information table, and global intention tracking is performed.
The embodiment of the application also provides a system for recommending products, which is applied to an intelligent dialogue system, and fig. 7 is a block diagram of a system for recommending products according to a sixth embodiment of the application, as shown in fig. 7, where the system includes:
the obtaining module 701 is configured to obtain dialogue content, extract dialogue features in the dialogue content, and obtain client static features in the intelligent dialogue system, and splice the dialogue features and the client static features;
a recall module 702, configured to input the splicing result into a classification model, where the classification model recalls a plurality of first candidate products, and when training the classification model, trains the classification model according to dialogue features and product selection results in the historical dialogue content, and according to the static features of the client;
the determining module 703 is configured to rank the plurality of first candidate products according to a preset product recommendation policy, and determine recommended products according to the product ranks.
In some of these embodiments, the dialog features include intent classification results, and in the acquisition module 701, the process of extracting dialog features in the dialog content includes:
dividing dialogue content into a plurality of clauses, determining whether the clauses exist in a question-format list, and if so, determining an intention classification result of the clauses according to the question-format list;
if not, inputting clauses into the classification model, and outputting the intention classification result of the clauses by the classification model.
In some of these embodiments, the intent classification results include the product type and product attributes of the customer consultation;
the recall module 702 is further configured to determine, after extracting the dialogue feature in the dialogue content, a product name corresponding to the dialogue feature according to the dialogue feature and according to the association relationship between the prestored product type, product name and product attribute, so as to obtain a plurality of second candidate products;
the determining module 703 is further configured to aggregate the first candidate product and the second candidate product to obtain a total set of candidate products; and sorting the total set of the plurality of candidate products according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
In some of these embodiments, the acquisition module 701 is further configured to:
after the dialogue content is acquired, dialogue content information is extracted in a dialogue round, and the information is filled into a preset slot of a global dialogue information table;
before determining whether the clause exists in the list of the question pattern, determining whether the clause structure is complete, if not, completing the clause according to the information of the history turns in the global information table.
It should be understood by those skilled in the art that the technical features of the above-described embodiments may be combined in any manner, and for brevity, all of the possible combinations of the technical features of the above-described embodiments are not described, however, they should be considered as being within the scope of the description provided herein, as long as there is no contradiction between the combinations of the technical features.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (8)
1. A method for recommending products, applied to an intelligent dialogue system, the method comprising:
acquiring dialogue content, extracting dialogue characteristics in the dialogue content, acquiring client static characteristics in the intelligent dialogue system, and splicing the dialogue characteristics and the client static characteristics;
inputting a splicing result into a classification model, wherein the classification model recalls a plurality of first candidate products, and training the classification model according to dialogue characteristics and product selection results in historical dialogue contents and according to the static characteristics of the clients when training the classification model;
sorting the plurality of first candidate products according to a preset product recommendation strategy, and determining recommended products according to the product sorting;
the dialogue features include intent classification results including product types and product attributes of customer consultations, and after extracting the dialogue features in the dialogue content, the method includes:
determining product names corresponding to the dialogue features according to the dialogue features and the association relation of prestored product types, product names and product attributes to obtain a plurality of second candidate products;
summarizing the first candidate product and the second candidate product to obtain a total set of candidate products;
and sorting the plurality of candidate product total sets according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
2. The method of claim 1, wherein the dialog features include intent classification results, and wherein the process of extracting dialog features in the dialog content comprises:
splitting the dialogue content into a plurality of clauses, determining whether the clauses exist in a question-format list, and if so, determining an intention classification result of the clauses according to the question-format list;
if not, inputting the clause into an intention classification model, and outputting an intention classification result of the clause by the intention classification model.
3. The method according to claim 2, characterized in that the method comprises:
after the dialogue content is acquired, dialogue content information is extracted in a dialogue round, and the information is filled into a preset slot of a global dialogue information table;
and before determining whether the clause exists in the list of question patterns, determining whether the clause structure is complete, and if not, completing the clause according to the information of the history turns in the global dialogue information table.
4. The method of claim 2, wherein after splitting the conversation content into a plurality of clauses, the method comprises, prior to determining whether the clause is present in a list of question patterns: extracting the entity in the clause, determining the entity type of the entity, and replacing the entity in the clause with the entity type.
5. The method of claim 4, wherein after the extracting the entity in the clause, the method further comprises: and determining whether the description of the entity meets a preset standard, and if not, changing the description of the entity into a standard description in a sentence.
6. A system for product recommendation, for use in an intelligent dialog system, the system comprising:
the acquisition module is used for acquiring dialogue content, extracting dialogue characteristics in the dialogue content, acquiring customer static characteristics in the intelligent dialogue system and splicing the dialogue characteristics and the customer static characteristics;
a recall module, configured to input a splice result into a classification model, where the classification model recalls a plurality of first candidate products, where the classification model is trained according to dialogue features and product selection results in historical dialogue content and according to the customer static features when the classification model is trained;
the determining module is used for sequencing the plurality of first candidate products according to a preset product recommendation strategy and determining recommended products according to the product sequencing;
the dialogue features include intent classification results including product types and product attributes of customer consultation;
the recall module is further configured to determine, after extracting the dialogue feature in the dialogue content, a product name corresponding to the dialogue feature according to the dialogue feature and according to a pre-stored association relationship between a product type, a product name and a product attribute, so as to obtain a plurality of second candidate products;
the determining module is further used for summarizing the first candidate product and the second candidate product to obtain a candidate product total set; and sorting the plurality of candidate product total sets according to a preset product recommendation strategy, and determining recommended products according to the product sorting.
7. The system of claim 6, wherein the dialog features include intent classification results, and wherein in the acquisition module, the process of extracting dialog features in the dialog content comprises:
splitting the dialogue content into a plurality of clauses, determining whether the clauses exist in a question-format list, and if so, determining an intention classification result of the clauses according to the question-format list;
if not, inputting the clause into an intention classification model, and outputting an intention classification result of the clause by the intention classification model.
8. The system of claim 7, wherein the acquisition module is further configured to:
after the dialogue content is acquired, dialogue content information is extracted in a dialogue round, and the information is filled into a preset slot of a global dialogue information table;
and before determining whether the clause exists in the list of question patterns, determining whether the clause structure is complete, and if not, completing the clause according to the information of the history turns in the global dialogue information table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110726350.6A CN113468420B (en) | 2021-06-29 | 2021-06-29 | Product recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110726350.6A CN113468420B (en) | 2021-06-29 | 2021-06-29 | Product recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113468420A CN113468420A (en) | 2021-10-01 |
| CN113468420B true CN113468420B (en) | 2024-04-05 |
Family
ID=77873738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110726350.6A Active CN113468420B (en) | 2021-06-29 | 2021-06-29 | Product recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113468420B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109146610A (en) * | 2018-07-16 | 2019-01-04 | 众安在线财产保险股份有限公司 | It is a kind of intelligently to insure recommended method, device and intelligence insurance robot device |
| CN109767318A (en) * | 2018-12-15 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Loan product recommended method, device, equipment and storage medium |
| CN110738545A (en) * | 2019-08-30 | 2020-01-31 | 深圳壹账通智能科技有限公司 | Product recommendation method and device based on user intention identification, computer equipment and storage medium |
| CN112115246A (en) * | 2020-08-14 | 2020-12-22 | 腾讯科技（深圳）有限公司 | Content recommendation method and device based on conversation, computer equipment and storage medium |
| CN112581203A (en) * | 2019-09-30 | 2021-03-30 | 微软技术许可有限责任公司 | Providing explanatory product recommendations in a session |
| CN112925892A (en) * | 2021-03-23 | 2021-06-08 | 苏州大学 | Conversation recommendation method and device, electronic equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9661067B2 (en) * | 2013-12-23 | 2017-05-23 | 24/7 Customer, Inc. | Systems and methods for facilitating dialogue mining |
| US10878479B2 (en) * | 2017-01-05 | 2020-12-29 | Microsoft Technology Licensing, Llc | Recommendation through conversational AI |
| US10932004B2 (en) * | 2017-01-24 | 2021-02-23 | Adobe Inc. | Recommending content based on group collaboration |
| US10762161B2 (en) * | 2017-08-08 | 2020-09-01 | Accenture Global Solutions Limited | Intelligent humanoid interactive content recommender |
| US10817667B2 (en) * | 2018-02-07 | 2020-10-27 | Rulai, Inc. | Method and system for a chat box eco-system in a federated architecture |
-
2021
- 2021-06-29 CN CN202110726350.6A patent/CN113468420B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109146610A (en) * | 2018-07-16 | 2019-01-04 | 众安在线财产保险股份有限公司 | It is a kind of intelligently to insure recommended method, device and intelligence insurance robot device |
| CN109767318A (en) * | 2018-12-15 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Loan product recommended method, device, equipment and storage medium |
| CN110738545A (en) * | 2019-08-30 | 2020-01-31 | 深圳壹账通智能科技有限公司 | Product recommendation method and device based on user intention identification, computer equipment and storage medium |
| CN112581203A (en) * | 2019-09-30 | 2021-03-30 | 微软技术许可有限责任公司 | Providing explanatory product recommendations in a session |
| WO2021066903A1 (en) * | 2019-09-30 | 2021-04-08 | Microsoft Technology Licensing, Llc | Providing explainable product recommendation in a session |
| CN112115246A (en) * | 2020-08-14 | 2020-12-22 | 腾讯科技（深圳）有限公司 | Content recommendation method and device based on conversation, computer equipment and storage medium |
| CN112925892A (en) * | 2021-03-23 | 2021-06-08 | 苏州大学 | Conversation recommendation method and device, electronic equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 一种基于对话的电子商务推荐系统;薛伟莲;王蕴慧;;辽宁师范大学学报(自然科学版)(02);全文 * |
| 中国移动智能客服系统研究及实现;胡珉;冯俊兰;王燕蒙;闪云香;;电信工程技术与标准化(10);全文 * |
| 对话式音乐推荐技术及系统实现;周纯伊;中国优秀硕士学位论文全文数据库;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113468420A (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10853582B2 (en) | Conversational agent | |
| US20210232761A1 (en) | Methods and systems for improving machine learning performance | |
| WO2020177282A1 (en) | Machine dialogue method and apparatus, computer device, and storage medium | |
| US20190272269A1 (en) | Method and system of classification in a natural language user interface | |
| US20180174037A1 (en) | Suggesting resources using context hashing | |
| CN114600099A (en) | Speech recognition accuracy enhancement using a natural language understanding-based meta-speech system of an assistant system | |
| US11482223B2 (en) | Systems and methods for automatically determining utterances, entities, and intents based on natural language inputs | |
| US10185983B2 (en) | Least-ask: conversational recommender system with minimized user interaction | |
| US11030251B2 (en) | Method and system for providing query suggestions based on personalized spelling correction | |
| WO2020206910A1 (en) | Product information pushing method, apparatus, computer device, and storage medium | |
| KR20170118983A (en) | Counseling system and method through robo advisor | |
| CN111767375A (en) | Semantic recall method and device, computer equipment and storage medium | |
| US20150348052A1 (en) | Crm-based discovery of contacts and accounts | |
| CN112650842A (en) | Human-computer interaction based customer service robot intention recognition method and related equipment | |
| CN113112282A (en) | Method, device, equipment and medium for processing consult problem based on client portrait | |
| CN113468420B (en) | Product recommendation method and system | |
| CN116521822B (en) | User intention recognition method and device based on 5G message multi-round session mechanism | |
| CN112989046B (en) | Real-time speech prejudging method, device, computer equipment and storage medium | |
| CN115481231A (en) | Session control method, device, computer equipment and storage medium | |
| WO2023034020A1 (en) | Sentence level dialogue summaries using unsupervised machine learning for keyword selection and scoring | |
| CN113297367A (en) | Method for generating user conversation linking language and related equipment | |
| AU2022204665B2 (en) | Automated search and presentation computing system | |
| US20230259541A1 (en) | Intelligent Assistant System for Conversational Job Search | |
| CN116525093A (en) | Prediction method, device, equipment and storage medium for session ending | |
| CN112597292A (en) | Question reply recommendation method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |