CN112966178B

CN112966178B - Consultation result distribution method, device, equipment and storage medium

Info

Publication number: CN112966178B
Application number: CN202110245111.9A
Authority: CN
Inventors: 郭冠军; 钟贤德; 袁冠红; 吕明
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2021-03-05
Filing date: 2021-03-05
Publication date: 2024-01-23
Anticipated expiration: 2041-03-05
Also published as: CN112966178A

Abstract

The application discloses a consultation result distribution method, a consultation result distribution device, consultation result distribution equipment and a storage medium, and relates to the technical field of artificial intelligence, in particular to the technical field of natural language processing and deep learning. One embodiment of the method comprises the following steps: if the predicted response probability of the user consultation information is larger than a first preset threshold value, distributing partial suppliers in the matched supplier set obtained based on the consultation information retrieval, wherein the distribution interval is a first preset duration; if the first preset time length is reached, determining whether the user consultation requirement is met, and distributing and matching the rest suppliers in the supplier set based on the determination result; and stopping distribution if the number of the distributed suppliers reaches a second preset threshold or the suppliers in the matched supplier set are distributed completely. The consultation result distribution method improves the distribution accuracy, reduces the occurrence probability that the user is excessively disturbed and the suppliers are excessively competing, and greatly improves the consultation experience of both the user and the suppliers.

Description

Consultation result distribution method, device, equipment and storage medium

Technical Field

The embodiment of the application relates to the field of computers, in particular to the field of artificial intelligence such as natural language processing, deep learning and the like, and particularly relates to a method, a device, equipment and a storage medium for distributing consultation results.

Background

In recent years, with the rapid development of the internet, forms are built on websites, so that users actively consult through the forms, and the form is a common way of consulting articles. By retrieving the consultation information entered by the user, many corresponding materials and suppliers can be recalled, and how to return matching suppliers to the user becomes an important issue because this directly affects the response rate of the user consultation information.

Disclosure of Invention

The embodiment of the application provides a method, a device, equipment and a storage medium for distributing consultation results.

In a first aspect, an embodiment of the present application provides a method for distributing a consultation result, including: calculating the expected responded probability of the user consultation information, and recording the expected responded probability as a first probability; distributing part of suppliers in the matched supplier set obtained by searching based on the user consultation information in response to the first probability being greater than a first preset threshold, wherein the distribution interval is a first preset duration; in response to reaching the first predetermined length of time, determining whether a customer's counseling appeal is satisfied, distributing a remaining portion of suppliers in the set of matching suppliers based on the determination; and stopping distribution in response to the number of distribution suppliers reaching a second preset threshold or the suppliers in the matched supplier set having been distributed.

In a second aspect, an embodiment of the present application provides a device for distributing a consultation result, including: a calculation module configured to calculate a probability that the user consultation information is expected to be responded, and record as a first probability; the first distribution module is configured to distribute partial suppliers in the matched supplier set obtained by searching based on the user consultation information in response to the first probability being greater than a first preset threshold value, wherein the distribution interval is a first preset duration; in response to reaching the first predetermined length of time, determining whether a customer's counseling appeal is satisfied, distributing a remaining portion of suppliers in the set of matching suppliers based on the determination; and a stopping module configured to stop distribution in response to the number of distribution suppliers reaching a second preset threshold or the suppliers in the matched supplier set having been distributed.

In a third aspect, an embodiment of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method as described in any one of the implementations of the first aspect.

In a fourth aspect, embodiments of the present application provide a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform a method as described in any implementation of the first aspect.

In a fifth aspect, embodiments of the present application propose a computer program product comprising a computer program which, when executed by a processor, implements a method as described in any of the implementations of the first aspect.

The method, the device, the equipment and the storage medium for distributing the consultation result, provided by the embodiment of the application, firstly calculate the expected response probability of the consultation information of the user and record the expected response probability as a first probability; then, distributing partial suppliers in the matched supplier set obtained by searching based on the user consultation information in response to the first probability being greater than a first preset threshold value, wherein the distribution interval is a first preset duration; in response to reaching the first predetermined length of time, determining whether a customer's counseling appeal is satisfied, distributing a remaining portion of suppliers in the set of matching suppliers based on the determination; and finally stopping distribution in response to the number of distribution suppliers reaching a second preset threshold or the distribution of suppliers in the matched supplier set. The application provides a consultation result distribution method, which effectively reduces the occurrence probability that users are disturbed by transition and suppliers are over-competing.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are for better understanding of the present solution and do not constitute a limitation of the present application. Wherein:

FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 is a flow chart of a first embodiment of a method of identifying a consultation intent according to the present application;

FIG. 3 is a flow chart of a second embodiment of a method of identifying a consultation intent according to the present application;

FIG. 4 is an exploded flowchart of the step of constructing a search expression of the recognition method of the counseling intents shown in FIG. 3;

FIG. 5 is a schematic diagram of an attribute tree constructed in accordance with a weight ordering;

FIG. 6 is a flow chart of a third embodiment of a method of identifying a consultation intent according to the present application;

FIG. 7 is an exploded flow chart of the step of constructing query terms of the method of recognizing a consultation intention shown in FIG. 6;

FIG. 8 is a flow chart of a first embodiment of a method of distributing consultation results according to the present application;

FIG. 9 is a flow chart of a second embodiment of a method of distributing consultation results according to the present application;

FIG. 10 is a flow chart of one implementation of a method of distributing consultation results according to the present application;

FIG. 11 is a flow chart of another implementation of a method of distributing consultation results according to the present application;

FIG. 12 is a schematic structural view of one embodiment of a distribution device of consultation results according to the present application;

fig. 13 is a block diagram of an electronic device for implementing a method of distributing consultation results of an embodiment of the present application.

Detailed Description

Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

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.

Fig. 1 illustrates an exemplary system architecture 100 to which embodiments of a method of distributing consultation results or a device of distributing consultation results of the present application may be applied.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or transmit information or the like. Various client applications, such as a browser or the like, may be installed on the terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices including, but not limited to, smartphones, tablets, laptop and desktop computers, and the like. When the terminal devices 101, 102, 103 are software, they can be installed in the above-described electronic devices. Which may be implemented as a plurality of software or software modules, or as a single software or software module. The present invention is not particularly limited herein.

The server 105 may provide various services. For example, the server 105 may analyze and process the consultation information acquired from the terminal devices 101, 102, 103 and generate processing results (e.g., vendor information).

The server 105 may be hardware or software. When the server 105 is hardware, it may be implemented as a distributed server cluster formed by a plurality of servers, or as a single server. When server 105 is software, it may be implemented as a plurality of software or software modules (e.g., to provide distributed services), or as a single software or software module. The present invention is not particularly limited herein.

It should be noted that, the method for distributing the consultation results provided in the embodiments of the present application is generally executed by the server 105, and accordingly, the device for distributing the consultation results is generally disposed in the server 105.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow 200 of a first embodiment of a method of identifying a consultation intent according to the present application is shown. The consultation intention recognition method comprises the following steps:

Step 201, filtering the title of the consulting article of the user and the pre-query word input by the user to obtain a first title and a first query word.

In this embodiment, the execution subject (e.g., the server 105 shown in fig. 1) of the recognition method of the counseling intents may filter the title of the counseling item of the user and the pre-query word input by the user to obtain the first title and the first query word. Filtering the title of the user consultation object to replace the word hit in the general stop word list and the text hit in the custom word set in the title of the user consultation object with a space, wherein the general stop word list can adopt the existing general stop word list, and the application is not particularly limited to the above; the custom word set is predefined, and may include text such as "manufacturer's direct sales", "high price, and" low price sales ". The preposed query words are query words input by a user before the consultation article is obtained, the preposed query words input by the user are filtered by judging whether the preposed query words of the user are company names, if the preposed query words are company names, the preposed query words are determined to be nonsensical query words, and the preposed query words are ignored; if not, the pre-query word is also filtered through the general stop word vocabulary, namely, the part of the pre-query word hitting the general stop word vocabulary is replaced by a space. The title of the user consulting article and the prepositioned query word input by the user are filtered, so that the filtered title and the filtered query word are obtained and recorded as a first title and a first query word.

Step 202, calculating the information coincidence degree of the first title and the first query word.

In this embodiment, after obtaining the first title and the first query term, the execution body may calculate the information overlap ratio of the first title and the first query term. The information overlap ratio of the first title and the first query term represents the matching degree and the relevance of the first title and the first query term. As an example, calculating the information overlap ratio of two text strings may be performed by calculating a ratio of the longest common sequence length of the two text strings to the length of the text string having the shorter length, and taking the ratio as the information overlap ratio of the short text string. In this embodiment, calculating the information overlap ratio of the first title and the first query word is to calculate the ratio of the longest common sequence length of the first title and the first query word to the length of the first query word, and take the ratio as the information overlap ratio of the first query word. The first title is obtained by filtering the title of the user consultation object, and the first query word is obtained by filtering the preposed query word input by the user, so that whether the preposed query word input by the user can accurately express the real consultation intention of the user can be judged by calculating the information coincidence degree of the first title and the first query word.

In some alternative implementations of this embodiment, TF-IDF values of two text strings may also be calculated by a pre-constructed word frequency-reverse document frequency (TF-IDF) model and formed into vectors, and then cosine similarity values of the two text strings may be calculated by using a cosine function, and simultaneously, the information amount of the short text may be determined by calculating a ratio of the length of the longest common sequence of the two text strings to the length of the text string having a shorter length as the information amount representative of the short text string, and by both values. That is, in this embodiment, TF-IDF values of two text strings of a first title and a first query word are calculated through a pre-constructed TF-IDF model and form a vector, then cosine similarity values of the two text strings are calculated by using a cosine function, and meanwhile, the information amount of a short text is determined through the two values together by calculating the ratio of the length of the longest common sequence of the two text strings to the length of the text string (the pre-query word) with a shorter length as the information amount representative of the short text string.

And 203, if the information coincidence degree is larger than a first preset threshold value, comparing the length of the first query word with a second preset threshold value, and determining the most complete query word based on the comparison result and searching.

In this embodiment, if the information overlap ratio is greater than a first preset threshold, the executing entity compares the length of the first query word with a second preset threshold, and determines and retrieves a full query word based on a comparison result, where the full query word is a word or phrase that can represent the most comprehensive consultation intention of the user. If the information overlap ratio calculated in step 202 is greater than a first preset threshold, it is indicated that the first query word can accurately express the consultation intention of the user. And then comparing the length of the first query word with a second preset threshold value, determining the most complete query word based on the comparison result, and comparing the length of the first query word with the second preset threshold value because the final search result range can be too large and inaccurate if the length of the first query word is too short, for example, the second preset threshold value can be set to be 2, judging whether the length of the first query word is less than 2, and determining the most complete query word based on the comparison result. And finally, searching based on the determined full query word.

In some optional implementations of the present embodiment, determining and retrieving the most complete query term based on the comparison result includes: if the length of the first query word is not smaller than the second preset threshold value, the first query word is used as the most complete query word for searching; if the length of the first query word is smaller than a second preset threshold value, the first title and the first attribute information of the pre-extracted consultation object are spliced to obtain second spliced information, and the second spliced information is used as the full query word to be searched. If the length of the first query word is not smaller than the second preset threshold value, the length of the first query word is proper, the search result meeting the consultation intention of the user can be obtained by searching based on the first query word, and the first query word can be used as the full query word for searching under the condition. If the length of the first query word is smaller than the second preset threshold value, the first query word is too short, and in this case, the first query word is used as the most complete query word to search, which may result in an excessively large and inaccurate search result range, so that a new most complete query word needs to be constructed at this time, and the first title and the first attribute information of the pre-extracted consulting article can be spliced to obtain second spliced information, where the first attribute information of the user consulting article may include, but is not limited to, brands, parameters, models, uses and the like of the article. And finally, searching the second spliced information as the most complete query word. Optionally, the user consults the identification of the article to access the pre-constructed abstract library, and extracts the corresponding title, the category, the corresponding label, the brand, the model, the parameter and other attribute information of the article.

In some optional implementations of this embodiment, if the information overlap ratio is not greater than a first preset threshold, the first attribute information of the pre-extracted consulting article and the first query word are spliced to obtain first spliced information, and the first spliced information is used as the most complete query word for searching. If the information overlap ratio calculated in step 202 is not greater than the first preset threshold, it is indicated that the overlap ratio between the first query word and the first header and the longest public sequence length of the first query word are low, which indicates that the first query word cannot comprehensively express the consultation intention of the user, so that the first attribute information of the pre-extracted consultation object and the first query word are spliced at this time, so as to obtain first spliced information, the first spliced information can more comprehensively express the consultation intention of the user, and finally the first spliced information is used as the most complete query word for searching.

According to the consultation intention recognition method provided by the embodiment of the application, firstly, the title of a consultation object of a user and a preposed query word input by the user are filtered to obtain a first title and a first query word; then calculating the information coincidence ratio of the first title and the first query word; and finally, if the information coincidence degree is larger than a first preset threshold value, comparing the length of the first query word with a second preset threshold value, and determining the most complete query word based on a comparison result and searching. The method for identifying the consultation intention can more accurately identify the consultation intention of the user, so that the most complete query word is generated and is searched based on the most complete query word, the accuracy of the matching result of the suppliers is further improved, the consultation experience of the user is enhanced, and the clue matching experience of the existing B2B platform is improved.

With continued reference to fig. 3, fig. 3 shows a flow 300 of a second embodiment of a method of identifying a consultation intent according to the present application. The consultation intention recognition method comprises the following steps:

step 301, searching based on the most complete query word to obtain recalled supplier information.

In this embodiment, the execution subject of the method for identifying a consultation intention may retrieve based on the most complete query word, thereby obtaining recalled vendor information. After constructing the most complete query word capable of expressing the user consultation intention, the executing body retrieves based on the most complete query word, and recalls the provider matched with the user consultation intention, thereby obtaining recalled provider information, wherein the recalled provider information can be the name of the recalled provider, the number of recalled providers and the like.

Step 302, in response to the number of recalled suppliers being less than a third preset threshold, constructing a retrieval expression containing a user consultation intention based on the pre-constructed word segmentation system and retrieving.

In this embodiment, in the case where the number of recalled suppliers is smaller than the third preset threshold, the execution subject may construct a search expression including the user consultation intention based on a pre-constructed word segmentation system and search. The most complete query word is a query word which expresses the consultation intention of the user to a high degree, if the number of the recalled suppliers based on the most complete query word is less than a third preset threshold value, a retrieval expression which contains the consultation intention of the user to a medium degree needs to be constructed, and the retrieval is carried out based on the retrieval expression so as to recall more materials and suppliers.

According to the consultation intention recognition method provided by the embodiment of the application, firstly, searching is carried out based on the most complete query word, and recalled supplier information is obtained; and then in response to the number of recalled suppliers being less than a third preset threshold, constructing a retrieval expression containing the user consultation intention based on the pre-constructed word segmentation system and retrieving. The application provides a consultation intention recognition method, which is used for searching based on the most complete query words capable of completely expressing the consultation intention of a user to obtain recalled provider information, wherein under the condition that the number of recalled providers is less than a preset threshold value, a search expression of the consultation intention of the user is contained in the construction process to a certain extent, and searching is carried out based on the search expression, so that more materials and providers are recalled, and further accuracy of matching results of the providers is improved.

With continued reference to fig. 4, fig. 4 shows a decomposition flow 400 of the step of constructing a retrieval expression of the recognition method of the counseling intents shown in fig. 3. The build search expression step may decompose the following:

and step 401, word segmentation is carried out on the most complete query words based on a pre-constructed word segmentation system, and word segmentation results are obtained.

In this embodiment, the word segmentation may be performed on the most complete query word based on a pre-constructed word segmentation system, so as to obtain a word segmentation result. The word segmentation system can be constructed by adopting the prior art, and the application is not particularly limited. The most complete query word may be a large spliced text, which may include a plurality of words, and the word segmentation system segments the most complete query word to return word segmentation results, and marks each word segmentation result as a term.

And step 402, splicing the first title, the second attribute information corresponding to the consultative object and the first query word to obtain third spliced information.

In this embodiment, the first title, the second attribute information corresponding to the consulting article, and the first query word may be spliced to obtain the third spliced information. The first title and the first query word are obtained by matching the title of the user consultation object with the pre-query word input by the user, and the second attribute information corresponding to the user consultation object may include, but is not limited to, a label corresponding to the user consultation object and a hosting range of a provider corresponding to the user consultation object, and the first title, the second attribute information corresponding to the consultation object and the first query word are spliced to obtain a complete spliced character string, and the complete spliced character string is recorded as third spliced information.

Step 403, counting word frequency of the word segmentation result in the third splicing information, and calculating word frequency-inverse text frequency index value of the word segmentation result.

In this embodiment, word frequency of each term in the third concatenation information is counted, and word frequency-inverse text frequency index value of the word segmentation result is calculated. TF-IDF is a common weighting technique used for information retrieval and text mining. TF-IDF is a statistical method that is generally used to evaluate the importance of a word to one of a set of documents or a corpus. As an example, the term returned by the term segmentation system may be traversed, and the term frequency of each term in the third concatenation information, that is, the frequency of each term in the third concatenation information, may be counted, and the reverse file frequency corresponding to the term may be queried, and finally the TF-IDF value of the term may be calculated, where TF-IDF is actually obtained by multiplying TF by IDF, and TF-IDF represents the weight of the term.

Step 404, selecting a predetermined number of word segmentation results based on the word frequency-inverse text frequency index value, and obtaining a word segmentation result set.

In this embodiment, a predetermined number of word segmentation results may be selected based on the calculated TF-IDF for each term, to obtain a word segmentation result set. As an example, TF-IDF values corresponding to each term may be ranked, and a predetermined number of terms with higher TF-IDF values may be selected to obtain a term set, e.g., the first six terms of TF-IDF values may be selected to obtain a term set.

Step 405, obtaining the search phrase based on the longest public matching sequence of the word segmentation result and the second attribute information in the word segmentation result set. .

In this embodiment, the term phrase may be obtained based on the longest common matching sequence of term in the term set and the second attribute information, where the second attribute information includes, but is not limited to, a label corresponding to the user consultation item and a camping scope of a provider corresponding to the user consultation item.

Step 406, constructing a search expression based on the search phrase, and searching based on the search expression.

In this embodiment, the term phrase obtained in step 405 is used as a search expression, and search is performed based on the search expression.

In some optional implementations of the present embodiment, if the term number is greater than 6, the term that is not the first six bits and is not attribute information such as region, article brand, article parameter, article signal, etc. is used to pre-form the optional word in the search expression; if the term number is not more than 6, the term which does not appear in the preposed query words such as region, article brand, article parameter, article signal and the like is used for dry pre-forming optional words in the retrieval expression. Taking the search expression with the completed intervention as a new search expression, and searching again based on the search expression.

In some optional implementations of this embodiment, when the most complete query word includes more attribute information (i.e., more term numbers) such as regions, article brands, article parameters, article signals, etc., if all attribute combinations are traversed according to the principle of "longest step-by-step decrease" and judged one by one, until the number of matched suppliers meets the requirement, the worst traversal number is: c (C) _n +C ¹ _n +C ² _n +……+C ⁿ _n ＝2 ⁿ Also of temporal complexity O (2 ⁿ ). This is unacceptable for a retrieval system. Therefore, the method provides a fastest searching algorithm, and the fastest searching algorithm can be utilized to search the search expression which can contain the consultation intention of the user, and the specific implementation process is as follows:

When the consultation intention of the user is found to contain a plurality of attributes such as A, B, C, D, E, firstly, all the attributes are ranked according to weight, and the ranking result is assumed to be: a > B > C > D, and then constructing an attribute tree according to the weight, wherein the attribute tree has the following definition:

(1) The root node is empty;

(2) Each node, except the root node, contains a character representation attribute;

(3) From the root node to a certain node, the characters passing through the path are connected to form a character string corresponding to the node, namely, attribute combination;

(4) Each character is added with a distinguishing ending symbol in the process of building a tree, which indicates that the paths from the root to each node are all a combination;

(5) Each node is an ending node, i.e. can stop from the root node to any node, which expresses the attribute combination corresponding to all nodes of the path through the path.

The attribute tree constructed in accordance with the weight ordering is shown in fig. 5, and fig. 5 is a schematic diagram of the attribute tree constructed in accordance with the weight ordering. It can be seen that the root node of the attribute tree shown in fig. 5 is empty and each node contains one character representation attribute, as shown in A, B, C, D. The problem of building the user consultation intention can then be converted into searching the longest maximum weight path meeting the requirement of the recall number of the retrieval scene in the attribute tree. The path finding algorithm comprises the following steps:

(1) Combining every second attribute, as AB, AC, AD, BC, BD, CD in fig. 5;

(2) Retrieving and recalling all the combinations, if the materials are recalled, marking the pair as 1, otherwise marking the pair as 0;

(3) Traversing pruning the attribute tree, including: traversing all the pairs marked as 0, pruning subtrees prefixed by the pair in the attribute tree, for example, if BC is not recalled by the combination, marking BC as 0, and pruning subtrees prefixed by BC as shown in fig. 5;

(4) The leftmost longest path, e.g., a- > B- > D, is found in the pruned tree.

The fastest find algorithm in this embodiment requires only C ² _n Step search operation, timeThe inter-complexity is O (n) ² ) That is, the fastest finding algorithm takes the time complexity from O (2 ⁿ ) Down to O (n) ² ). Under the scene with more attributes term, the searching effect and efficiency are improved.

Firstly, word segmentation is carried out on the most complete query words based on a pre-built word segmentation system to obtain word segmentation results; then splicing the first title, second attribute information corresponding to the consultation object and the first query word to obtain third spliced information; counting word frequency of the word segmentation result in the third splicing information, and calculating word frequency-inverse text frequency index value of the word segmentation result; then selecting a preset number of word segmentation results based on word frequency-inverse text frequency index values to obtain a word segmentation result set; obtaining a search phrase based on the longest public matching sequence of the word segmentation result and the second attribute information in the word segmentation result set; and finally, constructing a retrieval expression based on the retrieval phrase, and retrieving based on the retrieval expression. The method for constructing the search expression can be based on the search expression which contains the consultation intention of the user to a certain extent in the construction of the pre-constructed word segmentation system, and the search is performed based on the search expression so as to recall more materials and suppliers, so that the accuracy of the matching result of the suppliers is improved.

With continued reference to fig. 6, fig. 6 is a flow 600 of a third embodiment of a method of identifying a consultation intent according to the present application. The consultation intention recognition method comprises the following steps:

and step 601, calculating the number of materials meeting the consultation intention of the user, which are retrieved and recalled based on the most complete query words, and recording the number of materials as a first material number.

In this embodiment, the number of materials meeting the user's intention of consulting, which are retrieved based on the most complete query word, is calculated and recorded as the first number of materials. As an example, attribute information such as category, parameter, brand, region, model, usage, etc. of retrieving recall materials based on the most complete query word may be compared with the user query word to obtain common attributes of the two, and the number of the common attributes may be counted. If the attribute information is contained in the query words of the user, the recall material also contains the attribute information, and the matching is recorded as one time. If the value of the matching times/the total number of the common attributes is larger than a preset threshold value a, the matching times/the total number of the common attributes is recorded as materials meeting the user consultation intention, the materials are retrieved and recalled based on the most complete query words, and finally the total number of the materials meeting the user consultation intention, namely the first material number, which is the material number completely meeting the user consultation intention, is counted.

Step 602, calculating the material number meeting the consultation intention of the user for carrying out the retrieval recall based on the retrieval expression, and recording the material number as a second material number.

In this embodiment, the number of materials satisfying the user's intention of consultation, which is recalled based on the retrieval expression, is calculated and recorded as the second number of materials. As an example, attribute information such as category, parameter, brand, region, model, usage, etc. of the recall material retrieved based on the retrieval expression may be compared with the user query term to obtain a common attribute of the two, and the number of the common attributes may be counted. If the attribute information is contained in the query words of the user, the recall material also contains the attribute information, and the matching is recorded as one time. If the value of the matching times/the total number of the common attributes is larger than the preset threshold b, the matching times/the total number of the common attributes is recorded as materials meeting the user consultation intention for carrying out the retrieval recall based on the retrieval expression, and finally the total number of the materials meeting the user consultation intention for carrying out the retrieval recall based on the retrieval expression is counted, namely, the second material number is the material number basically meeting the user consultation intention.

And 603, comparing the sum of the first material number and the second material number with a fourth preset threshold value, and determining whether to construct a query word containing basic consultation requirements of the user based on the comparison result.

In this embodiment, the sum of the first material number and the second material number is compared with a fourth preset threshold, and whether to construct a query word containing the basic consultation appeal of the user is determined based on the comparison result. And calculating the sum x+y of the material number x which completely meets the user consultation intention and the material number y which basically meets the user consultation intention, comparing the value of x+y with a fourth preset threshold value, and determining whether the search result meets the consultation requirement of the user or not based on the comparison result, namely whether to perform the subsequent operation or not, namely whether to construct a query word containing the basic consultation requirement of the user.

In some optional implementations of this embodiment, if the sum of the first material count and the second material count is not greater than a fourth preset threshold, a query term including a user basic consultation appeal is constructed based on the weight value of the term result. The sum of the first material number and the second material number is not larger than a fourth preset threshold value, and the sum of the first material number and the second material number is indicated to be incapable of meeting the consultation appeal of the user, and at the moment, query words expressing the consultation appeal of the user to a low degree, namely query words containing the basic consultation appeal of the user, are required to be constructed.

In some optional implementations of this embodiment, if the sum of the first material count and the second material count is greater than a fourth preset threshold, no query term is constructed that includes the user's basic counseling appeal. The sum of the first material number and the second material number is larger than a fourth preset threshold value, which indicates that the sum of the first material number and the second material number can meet the consultation requirements of the user, and the construction of the query word is not needed.

If the query term is constructed, the search is performed based on the query term, step 604.

In this embodiment, if a query term is constructed, a search is performed based on the query term. Under the condition that a query word containing basic consultation complaints of a user is constructed, searching is carried out based on the query word so as to obtain more recall materials, and the recall result is combined with the result of searching recall based on the most complete query word and the searching expression so as to be used as a final recall result.

Step 605, retrieve recall providers based on the query terms and distribute recall providers to users.

In this embodiment, the recall provider and the material are retrieved based on the query term, and the execution subject may distribute the recalled provider to the user. As an example, matching suppliers may be distributed to users in batches in a time series.

According to the method for identifying the consultation intention, firstly, the number of materials meeting the consultation intention of the user and searched based on the most complete query words is calculated and recorded as the first material number; then calculating the material number meeting the consultation intention of the user for carrying out the retrieval recall based on the retrieval expression, and recording the material number as a second material number; comparing the sum of the first material number and the second material number with a fourth preset threshold value, and determining whether to construct a query word containing basic consultation requirements of the user based on a comparison result; if the query word is constructed, searching is carried out based on the query word; and finally searching the recall provider based on the query words and distributing the recall provider to the user. The method constructs a query word containing basic consultation requirements of a user to a low degree under the condition that the number of materials which completely meet the consultation requirements of the user and the number of materials which basically meet the consultation requirements of the user are smaller than a preset threshold value, retrieves the query word again based on the query word so as to recall more materials, combines the recall materials with a retrieval recall result based on the most complete query word and a retrieval expression to serve as a final recall result, so that the recall result can meet the consultation requirements of the user, the accuracy of the matching result of suppliers is improved, the consultation experience of the user is improved, and the clue matching experience of the existing B2B platform is further improved.

With continued reference to FIG. 7, FIG. 7 illustrates a decomposition flow 700 of the step of constructing query terms of the method of identifying a query intent shown in FIG. 6. The build query term step may decompose the following:

step 701, traversing the word segmentation weight of the word segmentation result to obtain a word segmentation result set larger than a preset weight threshold.

In this embodiment, the word segmentation weight of each term is traversed to obtain a word segmentation result set greater than a preset weight threshold. The word segmentation weight is returned after the word segmentation system performs word segmentation on the most complete query word. As an example, the term weight of each term may be traversed to obtain a term that is greater than a preset weight threshold and belongs to the term phrase set, so as to obtain a term result set, where the length of the term that is reserved needs to be greater than 1.

Step 702, adding the word segmentation weight of each word segmentation result in the word segmentation result set to the calculated weight to obtain the word segmentation result with the maximum weight value.

In this embodiment, the word segmentation weight of each term in the word segmentation result set is added to the calculation weight to obtain the word segmentation result with the largest weight value, where the calculation weight is a word frequency-inverse text frequency index value. In addition, if there are two or more bondable term, the weights of the respective bondable term are added and summed, and the sum is taken as the weight value of the term after bonding.

And step 703, taking the word segmentation result with the maximum weight value as a query word containing the basic consultation appeal of the user.

In this embodiment, term with the highest weight value in the calculation result of step 702 is taken as the query term containing the basic counseling appeal of the user.

Firstly traversing word segmentation weights of word segmentation results to obtain a word segmentation result set larger than a preset weight threshold; then adding the word segmentation weight of each word segmentation result in the word segmentation result set with the calculated weight to obtain a word segmentation result with the maximum weight value; and finally, taking the word segmentation result with the maximum weight value as a query word containing basic consultation appeal of the user. The method for constructing the query word comprises the steps of taking the word segmentation weight and the maximum word segmentation result of the calculation weight as the query word containing basic consultation appeal of a user, and searching again based on the query word to obtain a recall material, so that the material meeting the consultation requirement of the user can be recalled to the greatest extent.

With continued reference to fig. 8, fig. 8 shows a flow 800 of a first embodiment of a method of distributing consultation results according to the present application. The method for distributing the consultation result comprises the following steps:

Step 801, calculating the expected response probability of the user consultation information, and recording the expected response probability as a first probability.

In this embodiment, the execution body of the consultation result distribution method may calculate the probability that the user consultation information is expected to be responded, and record it as the first probability. The first probability is the probability that the information of the user consulting an item is expected to be responded to by the provider. As an example, the probability that the user consultation information is expected to be responded by the provider, i.e., the first probability, may be obtained by extracting the attribute information of the provider and the attribute information of the pre-query word input by the user online, substituting the extracted information into a pre-fitted multiple nonlinear statistical regression model, and recording it as P. The fitting process of the multi-element nonlinear statistical regression model can analyze corresponding data of historical consultation through offline, so that a mathematical model is built, and mathematical optimization is performed by using a least square method, so that a multi-element nonlinear statistical regression model is fitted. The historical consultation corresponding data comprises attribute information corresponding to the user consultation items of the provider attribute information, and the provider attribute information comprises: the sum of vendor response rates of the first five bits of historical response rate (top 5), the sum of vendor response rates of historical response rate top20, the number of high intent matching vendors based on the recall of the most complete query term, and the number of matching vendors in the matching vendor set; the attribute information corresponding to the user consulting article comprises: the method comprises the steps of historical consultation response rate under the same three-level industry with a user consultation object, historical consultation response rate under the same two-level industry with the user consultation object, historical consultation response rate under the same level industry with the user consultation object, the number of user names of the consultation user, the consultation response rate of top5 to which the user-specified consultation object belongs, the consultation historical response rate of the same receiving region (province) and the consultation historical response rate of the same receiving region (city), wherein the number of the user names of the consultation user can be 1 word, 2 words, 3 words or other. For example: assuming that the user consults that the article is a "steamed stuffed bun machine", the article belongs to the article under the third-level industry, the corresponding second-level industry is a "wheaten food machine", and the corresponding first-level industry is a "machine".

In some optional implementations of the present embodiment, calculating the probability that the user advisory information is expected to be responded to includes: and inputting the first attribute information corresponding to the consultation information and the second attribute information corresponding to the matched provider set into a pre-constructed nonlinear regression model to obtain the expected response probability of the consultation object. The first attribute information corresponding to the consulting information is attribute information corresponding to the consulting article of the user, and may include: the historical consultation response rate of the same class of industry, the same class of industry and the same class of industry with the consultation items of the user, the number of user names of the consultation users, the consultation response rate of the top5 to which the consultation items are assigned by the user, and the consultation historical response rate of the same receiving region (province and city). The matching provider set may be obtained by searching based on the most complete query word, the search expression and the query word containing the basic consultation appeal of the user, where the construction method of the most complete query word, the search expression and the query word containing the basic consultation appeal of the user corresponds to the foregoing embodiment, and the specific implementation may be described in the foregoing embodiment and will not be repeated herein.

Step 802, distributing partial suppliers in the matched supplier set obtained by searching based on the user consultation information in response to the first probability being greater than a first preset threshold, wherein the distribution interval is a first preset duration; in response to reaching the first predetermined length of time, determining whether a customer's counseling appeal is satisfied, distributing the remaining suppliers in the matching supplier set based on the determination.

In this embodiment, the executing entity distributes a portion of the suppliers in the matched supplier set when the first probability is greater than a first preset threshold, and the distribution interval is a first predetermined duration. The matched provider set is obtained by constructing the most complete query word by the previous value query word input by the user and then searching recall providers based on the most complete query word. And under the condition that the first probability P is larger than a preset threshold, distributing part of suppliers, wherein the distribution interval is a first preset time, if the first preset time is reached, determining whether the consultation requirement of the user is met or the quotation times reach the threshold, distributing the rest of suppliers in the matched supplier set based on the determination result, if the consultation requirement of the user is met or the quotation times reach the threshold, not distributing any more, and if the consultation requirement of the user is not met or the quotation times do not reach the threshold, continuing to distribute the rest of suppliers in the matched supplier set. If the provider of the second distribution still fails to meet the customer's consultation appeal, the provider continues to be distributed.

In step 803, the distribution is stopped in response to the number of distribution suppliers reaching a second preset threshold or the suppliers in the set of matching suppliers having been distributed.

In this embodiment, when the number of distributed suppliers reaches the second preset threshold or the suppliers in the matched supplier set have been distributed, the distribution is stopped. The number of suppliers that are distributed via step 802 reaching the second preset threshold indicates that enough suppliers have been distributed to the consulting user, at which point no more distribution is performed to prevent the user from receiving too much supplier information to be disturbed or when the suppliers in the matched set of suppliers have been distributed.

In some alternative implementations of this embodiment, after distributing the matching provider, the advisory information is placed in the passive distribution unit if no one responds within the predicted longest response time of the advisory information. The consultation information in the passive distribution unit can be seen by suppliers in the same industry preferentially, and the suppliers can be contacted for quotation when the suppliers which are matched with the consultation intention of the user to the highest degree have not been quoted.

According to the consultation result distribution method provided by the embodiment of the application, firstly, the probability that the consultation information of the user is expected to be responded is calculated and recorded as the first probability; then, distributing partial suppliers in the matched supplier set obtained by searching based on the preposed query words input by the user in response to the first probability being greater than a first preset threshold, wherein the distribution interval is a first preset duration; in response to reaching the first predetermined length of time, determining whether a customer's counseling appeal is satisfied, distributing a remaining portion of suppliers in the set of matching suppliers based on the determination; and finally stopping distribution in response to the number of distribution suppliers reaching a second preset threshold or the distribution of suppliers in the matched supplier set. The method for distributing the consultation results for the users is characterized in that the responding probability of the consultation information of the users is estimated, the matched suppliers are distributed for the users batch by batch according to the time sequence based on the estimated responding probability, timeliness and accuracy of platform clue distribution are improved, the occurrence probability that the users are excessively disturbed and the suppliers are excessively contended is effectively reduced, and consultation experience of the users and the suppliers is greatly improved.

With continued reference to fig. 9, fig. 9 shows a flow 900 of a second embodiment of a method of distributing consultation results according to the present application. The method for distributing the consultation result comprises the following steps:

step 901, traversing the recalled materials retrieved based on the user consultation information, and extracting the supplier information corresponding to the materials.

In this embodiment, the execution body of the method for distributing the consultation result may traverse the recalled material retrieved based on the user consultation information, and extract the provider information corresponding to the material. The full query word and the search expression can be constructed based on the pre-query word input by the user and the attribute information of the user consultation object, the complete matching material can be recalled based on the full query word, the basic matching material can be recalled based on the search expression, and the supplier information corresponding to all the materials is extracted by traversing the complete matching material and the basic matching material.

Step 902, sorting suppliers based on a pre-constructed sorting model to obtain a matched supplier set.

In this embodiment, the executing body may rank the suppliers obtained in step 901 based on a pre-constructed ranking model, so as to return a ranked set of matched suppliers. Learning To Rank (LTR) is a supervised Learning ranking method. Inputting the complete matching materials, the basic matching materials and the corresponding supplier information into a sorting model, sorting and de-duplicating the suppliers based on the returned scores and the number of the suppliers hit to the matching materials (complete matching materials/basic matching materials) and other factors, and thus obtaining a final matching supplier set.

Step 903, calculating the expected response probability of the user consultation information, and calculating the total number of suppliers with the sum of the historical response rates as a predetermined value.

In this embodiment, the execution body may calculate the probability that the user consultation information is expected to be responded to, and calculate the total number of suppliers whose sum of the historical response rates is a predetermined value. The calculated probability of the user consulting information being expected to be responded in this step corresponds to step 801 of the foregoing embodiment, and the specific implementation may refer to the foregoing description of step 801, which is not repeated herein.

The total number of suppliers whose sum of the historical response rates is a predetermined value may be calculated by the offline extracted supplier attribute information. As an example, a total number of suppliers with a sum of the historical response rates of 300% may be calculated by the supplier attribute information extracted offline, which is denoted as N.

In step 904, in response to the first probability being greater than a first preset threshold, distributing part of the suppliers in the matched supplier set retrieved based on the user consultation information, wherein the distribution interval is a first predetermined time period, and in response to reaching the first predetermined time period, determining whether the consultation requirement of the user is satisfied, and distributing the rest of the suppliers in the matched supplier set based on the determination result.

In this embodiment, if the first probability P is greater than a first preset threshold, distributing part of the suppliers in the matched supplier set, wherein the distribution interval is a first predetermined duration, and if the first predetermined duration is reached, determining whether the consultation requirement of the user is satisfied, and distributing the rest of the suppliers in the matched supplier set based on the determination result. Step 904 corresponds to step 802 of the foregoing embodiment, and the specific implementation may refer to the foregoing description of step 802, which is not repeated herein.

In some optional implementations of the present embodiment, distributing a portion of the suppliers in the set of matched suppliers retrieved based on the user consultation information includes: a first number of suppliers in the set of matched suppliers is distributed based on the historical response rate. That is, if P is greater than the first preset threshold, the head suppliers with higher historical response rates in the matched supplier set are distributed, the number of the distributed suppliers is N, and the distribution interval is the first predetermined duration T. And distributing a preset number of suppliers with high historical response rate to the users, so that the distributed suppliers can respond to the consultation demands of the users as soon as possible, and the response rate of the consultation is improved to the greatest extent.

In some alternative implementations of the present embodiment, the distribution of the set of matching suppliers is responsive to reaching the first predetermined length of time and the user's counseling appeal not being metAnd a second number of suppliers remain, wherein the distribution interval is a second predetermined time length, the second number is larger than the first number, and the second predetermined time length is smaller than the first predetermined time length. If the consultation requirement of the user is not satisfied or the quotation times of the suppliers do not reach the threshold value within the time T, continuously distributing the suppliers for the user, distributing the rest suppliers in the matched supplier set, wherein the distribution number is k ₁ * N, where k ₁ > 1, distribution interval k ₁ * T, where k ₂ <1. The number of suppliers for the second distribution is larger than that of suppliers for the first distribution, and the time interval of the second distribution is smaller than that of the first distribution, so that the response rate of the user consultation is improved. If after the second distribution of suppliers, the customer's counseling appeal is still not met or the number of supplier invoices is still not up to a threshold, then the loop continues to step 904, continuing to distribute the remaining suppliers in the matched supplier set to the customer.

In some alternative implementations of the present embodiment, the dispensing is stopped in response to reaching the first predetermined length of time and the user's counseling appeal being met. If the consultation requirement of the user is satisfied or the quotation times of the suppliers reach the threshold value within the time T, the distribution of the suppliers is stopped.

Step 905, distributing a third number of suppliers in the matched supplier set based on the historical response rate in response to the first probability not being greater than the first preset threshold, the distribution interval being a third predetermined duration; in response to reaching the third predetermined length of time, determining whether the customer's counseling appeal is satisfied, distributing the remaining suppliers in the matching supplier set based on the determination.

In this embodiment, in the case where the first probability P is not greater than the first preset threshold, then a third number of suppliers in the matched supplier set, that is, the head suppliers with high historical response rates in the matched supplier set are distributed based on the historical response rates, the distribution number is W ₁ ，W ₁ ＝k ₃ * N, where k ₃ More than 1, distribution interval is T ₁ ，T ₁ ＝T/K ₃ . In response to reaching the third predetermined length of time,determining whether the consultation appeal of the user is satisfied, distributing the rest of suppliers in the matched supplier set based on the determination result, if the third preset time is reached, but the consultation appeal of the user is not satisfied or the quotation times are not reached to the threshold value, continuing to distribute the rest of suppliers in the matched supplier set, otherwise stopping distribution. In addition, if the provider of the second distribution still fails to meet the customer's counseling appeal, the provider continues to be distributed.

In some alternative implementations of the present embodiment, in response to reaching the third predetermined length of time and the consultation appeal of the user not being met, a remaining fourth number of suppliers in the set of matching suppliers are distributed with a distribution interval of a fourth predetermined length of time, wherein the fourth number is greater than the third number and the fourth predetermined length of time is less than the third predetermined length of time. If the consultation requirement of the user is not met or the quotation times of the suppliers do not reach the threshold value for the third preset time period, continuously distributing the suppliers for the user, distributing the rest suppliers in the matched supplier set, wherein the distribution number is W ₂ ，W ₂ ＝k ₃ *W ₁ The distribution interval is T ₂ ，T ₂ ＝T ₁ /k ₃ . The number of suppliers for the second distribution is larger than that of suppliers for the first distribution, and the time interval of the second distribution is smaller than that of the first distribution, so that the response rate of the user consultation is improved.

In some alternative implementations of the present embodiment, the dispensing is stopped in response to reaching a third predetermined length of time and the user's counseling appeal being met. If T ₁ And in the time, if the consultation requirement of the user is met or the quotation times of the provider reach the threshold value, stopping the distribution of the provider.

In some alternative implementations of the present embodiment, the third number is greater than the first number and the third predetermined time period is less than the first predetermined time period. And when P is not more than the first preset threshold, the number of suppliers distributed is more than the number of suppliers distributed when P is more than the first preset threshold, and the time interval of the suppliers distributed when P is not more than the first preset threshold is less than the time interval of the suppliers distributed when P is more than the first preset threshold, so that more suppliers are distributed to users in a shorter time interval when the response probability of the user consultation information by the suppliers is small, and the response rate of the user consultation information is improved.

And step 906, stopping distribution in response to the number of distributed suppliers reaching a second preset threshold or the suppliers in the matched supplier set having been distributed.

In this embodiment, when the number of distributed suppliers reaches the second preset threshold or the suppliers in the matched supplier set have been distributed, the distribution is stopped. Step 906 corresponds to step 803 of the foregoing embodiment, and specific implementation may refer to the foregoing description of step 803, which is not repeated herein.

According to the consultation result distribution method provided by the embodiment of the application, firstly, materials recalled based on user consultation information are searched and retrieved, and supplier information corresponding to the materials is extracted; sequencing suppliers based on a pre-constructed sequencing model to obtain a matched supplier set; then calculating the expected responded probability of the user consultation information, and calculating the total number of suppliers with the sum of the historical response rates as a preset value; then, in response to the first probability being greater than a first preset threshold, distributing part of suppliers in the matched supplier set obtained by searching based on the pre-query words input by the user, wherein the distribution interval is a first preset time length, in response to the first preset time length being reached, determining whether the consultation requirements of the user are met, and distributing the rest of suppliers in the matched supplier set based on the determination result; distributing a third number of suppliers in the matched set of suppliers based on the historical response rate in response to the first probability not being greater than a first preset threshold, the distribution interval being a third predetermined length of time; in response to reaching the third predetermined length of time, determining whether the customer's counseling appeal is satisfied, distributing the remaining suppliers in the matching supplier set based on the determination; and finally stopping distribution in response to the number of distribution suppliers reaching a second preset threshold or the distribution of suppliers in the matched supplier set. The method for distributing the consultation results for the users distributes matched suppliers for the users in batches according to the time sequence, when the distributed suppliers do not meet the consultation requirements of the users, the distribution interval is shorter, the number of the distributed suppliers is more, the effect of rapidly matching the proper suppliers for the users is achieved, timeliness and accuracy of platform clue distribution are improved, the occurrence probability that the users are excessively disturbed and the suppliers are excessively contended is effectively reduced, and consultation experience of the users and the suppliers is greatly improved.

With continued reference to fig. 10, fig. 10 is a flowchart showing one implementation of a method for distributing a consultation result according to the present application, as shown in fig. 10, for distributing a provider based on a recall of user consultation information to a user, after distributing the provider, judging whether a consultation appeal of the user is satisfied, and if the consultation appeal of the user is satisfied, stopping the distribution; if the consultation requirements of the user are not met, continuing to distribute the suppliers, after the second distribution, continuing to judge whether the consultation requirements of the user are met or whether the response times of the suppliers reach a threshold value, and if the consultation requirements of the user are met or the response times of the suppliers reach the threshold value, stopping distribution; if the consultation requirements of the users are not met or the response times of the suppliers do not reach the threshold value, the distribution steps of the suppliers are circulated until the number of the distributed suppliers reaches a second preset threshold value or the suppliers in the matched supplier set are distributed, and the distribution is stopped.

With continued reference to fig. 11, fig. 11 is another implementation flowchart of a method for distributing a consultation result according to the present application, as shown in fig. 11, firstly, a user inputs consultation information, a parsing and extracting unit extracts attribute information such as a title, a label, a category, a brand, a model, a parameter, etc. corresponding to a user consultation item, then a consultation full information query word constructing unit constructs a full query word capable of expressing a user consultation intention based on the extracted attribute information and the user consultation information, and then a query optimizing unit (query word optimizing unit), an expression rewriting unit (expression rewriting unit) and a query price reducing unit optimize the constructed full query word, thereby generating information expressing the user consultation intention to a high, medium and low degree, and searching recall matching materials and suppliers based on the generated query word capable of expressing the user consultation intention. And then, the user consultation attribute extraction unit and the supplier attribute extraction unit extract user consultation attribute information and supplier attribute information based on the user consultation information and the obtained matched supplier set, and input the extracted user consultation attribute information and supplier attribute information into a pre-constructed multiple regression model so as to obtain the probability that the user consultation information predicts the response of the supplier. And carrying out distribution of the suppliers based on the estimated responded probability, continuously comparing the estimated responded probability with a first preset threshold value, determining the number of the distributed suppliers and the distribution time interval based on the comparison result, and stopping distribution when the number of the distributed suppliers reaches a second preset threshold value or the suppliers in the matched supplier set are distributed. And the consultation experience of both the user and the provider is improved.

With further reference to fig. 12, as an implementation of the method shown in the foregoing figures, the present application provides an embodiment of a device for distributing consultation results, where the embodiment of the device corresponds to the embodiment of the method shown in fig. 2, and the device may be specifically applied to various electronic devices.

As shown in fig. 12, the consultation result distribution apparatus 1200 of the present embodiment may include: a calculation module 1201, a first distribution module 1202 and a stop module 1203. Wherein, the calculating module 1201 is configured to calculate a probability that the user consultation information is expected to be responded, and record as a first probability; a first distribution module 1202 configured to distribute, in response to the first probability being greater than a first preset threshold, a portion of the suppliers in the set of matched suppliers retrieved based on the user-advisory information at a distribution interval of a first predetermined length of time; in response to reaching the first predetermined length of time, determining whether a customer's counseling appeal is satisfied, distributing a remaining portion of suppliers in the set of matching suppliers based on the determination; a stopping module 1203 is configured to stop the distribution in response to the number of distribution suppliers reaching a second preset threshold or the suppliers in the set of matching suppliers having been distributed.

In the present embodiment, in the consultation result distribution apparatus 1200: the specific processes and technical effects of the calculation module 1201, the first distribution module 1202 and the stopping module 1203 may refer to the relevant descriptions of steps 201-203 in the corresponding embodiment of fig. 2, and are not repeated here.

In some optional implementations of this embodiment, the first distribution module is further configured to: a first number of suppliers in the set of matched suppliers is distributed based on the historical response rate.

In some optional implementations of this embodiment, the first distribution module is further configured to: distributing a remaining second number of suppliers in the set of matched suppliers at a second predetermined time interval in response to reaching the first predetermined time period and the consultation appeal of the user not being satisfied, wherein the second number is greater than the first number and the second predetermined time period is less than the first predetermined time period; and stopping distribution in response to reaching the first predetermined time period and the user's counseling appeal being satisfied.

In some optional implementations of this embodiment, the apparatus for distributing the consultation result further includes: a second distribution module configured to distribute a third number of suppliers in the matched set of suppliers based on the historical response rate in response to the first probability not being greater than the first preset threshold, the distribution interval being a third predetermined length of time; in response to reaching the third predetermined length of time, determining whether the customer's counseling appeal is satisfied, distributing the remaining suppliers in the matching supplier set based on the determination.

In some optional implementations of this embodiment, the second distribution module is further configured to: distributing a remaining fourth number of suppliers in the set of matched suppliers at a fourth predetermined time period in response to reaching the third predetermined time period and the consultation appeal of the user not being satisfied, wherein the fourth number is greater than the third number and the fourth predetermined time period is less than the third predetermined time period; and stopping distribution in response to reaching the third predetermined time period and the user's counseling appeal being satisfied.

In some alternative implementations of the present embodiment, the third number is greater than the first number and the third predetermined time period is less than the first predetermined time period.

In some optional implementations of this embodiment, the apparatus for distributing the consultation result further includes: the traversing module is configured to traverse the materials retrieved and recalled based on the user consultation information and extract the supplier information corresponding to the materials; the ranking module is configured to rank suppliers based on a pre-constructed ranking model to obtain a matched supplier set.

In some optional implementations of the present embodiment, the computing module is further configured to: and inputting the first attribute information corresponding to the consultation information and the second attribute information corresponding to the matched provider set into a pre-constructed nonlinear regression model to obtain the expected response probability of the consultation object of the user.

According to embodiments of the present application, there is also provided an electronic device, a readable storage medium and a computer program product.

Fig. 13 illustrates a schematic block diagram of an example electronic device 1300 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 13, the apparatus 1300 includes a computing unit 1301 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 1302 or a computer program loaded from a storage unit 1308 into a Random Access Memory (RAM) 1303. In the RAM 1303, various programs and data required for the operation of the device 1300 can also be stored. The computing unit 1301, the ROM 1302, and the RAM 1303 are connected to each other through a bus 1304. An input/output (I/O) interface 1305 is also connected to bus 1304.

Various components in device 1300 are connected to I/O interface 1305, including: an input unit 1306 such as a keyboard, a mouse, or the like; an output unit 1307 such as various types of displays, speakers, and the like; storage unit 1308, such as a magnetic disk, optical disk, etc.; and a communication unit 1309 such as a network card, a modem, a wireless communication transceiver, or the like. The communication unit 1309 allows the device 1300 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

The computing unit 1301 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 1301 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The calculation unit 1301 performs the respective methods and processes described above, for example, a distribution method of consultation results. For example, in some embodiments, the method of distributing the consultation results may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 1308. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 1300 via the ROM 1302 and/or the communication unit 1309. When the computer program is loaded into the RAM 1303 and executed by the computing unit 1301, one or more steps of the above-described consultation result distribution method may be performed. Alternatively, in other embodiments, computing unit 1301 may be configured to perform the distribution method of the consultation result in any other suitable manner (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A consultation result distribution method, comprising:

calculating the expected responded probability of the user consultation information, and recording the expected responded probability as a first probability;

distributing part of suppliers in the matched supplier set obtained by searching based on the user consultation information in response to the first probability being greater than a first preset threshold, wherein the distribution interval is a first preset time length, determining whether the consultation requirement of the user is met in response to the first preset time length, and distributing the rest of suppliers in the matched supplier set based on a determination result;

and stopping distribution in response to the distribution of the number of suppliers reaching a second preset threshold or the distribution of the suppliers in the matched set of suppliers being completed.

2. The method of claim 1, wherein the distributing of the portion of the set of matched suppliers retrieved based on the user consultation information includes:

A first number of suppliers in the set of matched suppliers are distributed based on historical response rates.

3. The method of claim 2, wherein the determining whether the customer's counseling appeal is satisfied in response to reaching the first predetermined length of time, distributing the remaining portion of the set of matched suppliers based on the determination comprises:

distributing a remaining second number of suppliers in the set of matched suppliers with a second predetermined time period in response to reaching the first predetermined time period and the user's counseling appeal not being satisfied, wherein the second number is greater than the first number and the second predetermined time period is less than the first predetermined time period;

stopping distribution in response to reaching the first predetermined length of time and the user's counseling appeal being satisfied.

4. The method of claim 3, wherein after the calculating the probability that the user-advisory information is expected to be responded to, the method further comprises:

and distributing a third number of suppliers in the matched supplier set based on historical response rates in response to the first probability not being greater than the first preset threshold, wherein a distribution interval is a third preset duration, determining whether consultation requirements of the user are met in response to reaching the third preset duration, and distributing the rest of suppliers in the matched supplier set based on a determination result.

5. The method of claim 4, wherein the determining whether the customer's counseling appeal is satisfied in response to reaching the third predetermined length of time, distributing the remaining portion of the set of matched suppliers based on the determination comprises:

distributing a remaining fourth number of suppliers in the set of matched suppliers at a fourth predetermined time period in response to reaching the third predetermined time period and the customer's counseling appeal not being satisfied, wherein the fourth number is greater than the third number and the fourth predetermined time period is less than the third predetermined time period;

stopping distribution in response to reaching the third predetermined length of time and the user's counseling appeal being satisfied.

6. The method of claim 5, wherein the third number is greater than the first number and the third predetermined time period is less than the first predetermined time period.

7. The method of any of claims 1 or 6, wherein prior to the calculating the probability that user-advisory information is expected to be responded to, the method further comprises:

traversing the recalled materials retrieved based on the user consultation information, and extracting provider information corresponding to the materials;

And sequencing the suppliers based on a pre-constructed sequencing model to obtain a matched supplier set.

8. The method of claim 7, the calculating the probability that user-advisory information is expected to be responded to, comprising:

and inputting the first attribute information corresponding to the consultation information and the second attribute information corresponding to the matched provider set into a pre-constructed nonlinear regression model to obtain the expected response probability of the consultation object of the user.

9. A consultation result distribution apparatus, comprising:

a calculation module configured to calculate a probability that the user consultation information is expected to be responded, and record as a first probability;

the first distribution module is configured to distribute partial suppliers in the matched supplier set obtained by searching based on the user consultation information in response to the first probability being greater than a first preset threshold, the distribution interval is a first preset duration, whether the consultation requirements of the user are met is determined in response to the first preset duration, and the rest of suppliers in the matched supplier set are distributed based on a determination result;

a stopping module configured to stop distribution in response to the number of distributed suppliers reaching a second preset threshold or the suppliers in the matched set of suppliers having been distributed.

10. The apparatus of claim 9, wherein the first distribution module is further configured to:

11. The apparatus of claim 10, wherein the first distribution module is further configured to:

12. The apparatus of claim 11, wherein the apparatus further comprises:

and a second distribution module configured to distribute a third number of suppliers in the matched supplier set based on a historical response rate in response to the first probability not being greater than the first preset threshold, the distribution interval being a third predetermined duration, determine whether a counseling appeal of the user is satisfied in response to reaching the third predetermined duration, and distribute the remaining suppliers in the matched supplier set based on the determination result.

13. The apparatus of claim 12, wherein the second distribution module is further configured to:

14. The apparatus of claim 13, wherein the third number is greater than the first number, and the third predetermined duration is less than the first predetermined duration.

15. The apparatus of any of claims 9-14, wherein the apparatus further comprises:

the traversing module is configured to traverse the materials retrieved based on the user consultation information and extract the supplier information corresponding to the materials;

and the ranking module is configured to rank the suppliers based on a pre-constructed ranking model to obtain a matched supplier set.

16. The apparatus of claim 15, wherein the computing module is further configured to:

17. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

18. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-8.