CN114896518A

CN114896518A - Recommendation method and device for business case, computer equipment and storage medium

Info

Publication number: CN114896518A
Application number: CN202210584684.9A
Authority: CN
Inventors: 高民; 熊嘉娜; 张洁; 肖地长
Original assignee: Ping An Bank Co Ltd
Current assignee: Ping An Bank Co Ltd
Priority date: 2022-05-26
Filing date: 2022-05-26
Publication date: 2022-08-12

Abstract

The embodiment of the application belongs to the field of big data and the technical field of artificial intelligence, and relates to a recommendation method of a business case, which comprises the steps of obtaining current business information of the current business case; extracting a sample business case group with at least one current label in a data asset library based on the current label; analyzing each sample business case in the sample business case group by applying NLP semantics; establishing a scoring matrix according to the current business information of the current business case, the sample business information of the sample business case and a pre-stored evaluation score, and calculating the similar dimension of the current business information of the current business case and the sample business information of the sample business case in the scoring matrix based on a collaborative filtering algorithm. The application also provides a recommendation device, computer equipment and storage medium for the business case. In addition, the application also relates to a block chain technology, and the service cases can be stored in the block chain. The method and the device complete the recommendation of the business case.

Description

Recommendation method and device for business case, computer equipment and storage medium

Technical Field

The present application relates to the field of big data and the technical field of artificial intelligence, and in particular, to a method and an apparatus for recommending a business case, a computer device, and a storage medium.

Background

The storage mode of the past business of a company is usually paper storage or only case detail recording storage on a computer, for the colleagues of a new company, the contents of each past business need to be browsed to be familiar with the past business, a long adaptation and familiarity process is needed for how to quickly familiarize the past business of the new company and accurately apply the experience of the past business to the development of the new business, the time consumed is increased and the proficiency of accurate application is lower as the business scale is enlarged and the business volume is increased and the contents to be familiar are increased.

Disclosure of Invention

An object of the embodiments of the present application is to provide a method and an apparatus for recommending a service case, a computer device, and a storage medium, so as to solve the technical problems that a newly-enrolled peer is low in efficiency of being familiar with a previous service, and the matching degree of a service is low when the new service is screened.

In order to solve the above technical problem, an embodiment of the present application provides a method for recommending a business case, which adopts the following technical solutions: the method comprises the following steps:

acquiring current service information of a current service case, wherein the current service information comprises at least one current tag;

extracting a sample business case group with at least one current label in a data asset library based on the current label, wherein the sample business case group comprises at least one sample business case;

analyzing each sample service case in the sample service case group by applying NLP semantics, and obtaining sample service information according to an analysis result;

establishing a scoring matrix according to the current business information of the current business case, the sample business information of the sample business case and a prestored evaluation score, and calculating the similar dimension of the current business information of the current business case and the sample business information of the sample business case in the scoring matrix based on a collaborative filtering algorithm;

acquiring a sample business case with the highest similarity dimension with the current business case as a target business case according to the scoring matrix;

and acquiring the target service information of the target service case, and recommending the target service case.

Further, the step of analyzing each sample service case in the sample service case group by applying NLP semantics and obtaining sample service information according to an analysis result includes:

analyzing a language structure in each sample service case in the sample service case group by applying NLP semantics to obtain sample service information, wherein the sample service information comprises sample service contents and sample service nodes;

and extracting key labels from the sample service contents, and sequencing the sample service contents according to the sample service nodes.

Further, the scoring matrix includes a current scoring submatrix and a sample scoring submatrix, the step of establishing the scoring matrix according to the current business information of the current business case, the sample business information of the sample business case and a pre-stored evaluation score, and the step of calculating the similar dimension of the current business information of the current business case and the sample business information of the sample business case in the scoring matrix based on a collaborative filtering algorithm includes:

acquiring the pre-stored evaluation score and the current service node of the current service case, and acquiring the pre-stored evaluation score and the sample service node of the sample service case;

splitting the pre-stored evaluation score of the current service case according to the current service node, acquiring the node score of at least one current service case, and establishing a current evaluation sub-matrix according to the node score of the current service case; splitting the pre-stored evaluation scores of the sample business case according to the sample business nodes, acquiring at least one node score of the sample business case, and establishing a sample evaluation molecular matrix according to the node scores of the sample business case;

and calculating the similar dimensionality of the current scoring submatrix and the sample scoring submatrix based on a collaborative filtering algorithm.

Further, the step of establishing a current scoring matrix according to the node scores of the current business case comprises:

taking the head of a row as the current service node and the head of a column as the name of the current service case, filling a node score corresponding to the current service node in a table, and establishing a current scoring submatrix, wherein the current service node is associated with the current service content;

the step of establishing the sample evaluation matrix according to the node scores of the sample business cases comprises the following steps:

and establishing a sample evaluation molecular matrix by taking the head of a row as the sample service node and the head of a column as the name of the sample service case and filling the node scores corresponding to the sample service node in a table, wherein the sample service node is related to the content of the sample service.

Further, the step of calculating the similar dimension of the current scoring submatrix and the sample scoring submatrix based on the collaborative filtering algorithm includes:

taking the product of the sum of the node scores of the current business case and the sum of the node scores of the sample business case as a dimension numerator, and taking the product of the sum of the node scores of the current business case squared and the sum of the node scores of the sample business case squared as a dimension denominator;

and dividing the dimension numerator by the dimension denominator to obtain a value as the similar dimension of the current business case and the sample business case.

Further, after the step of obtaining the target service information of the target service case and recommending the target service case, the method further includes:

extracting reason guide factors of each key concern case node of the target business case, which is successfully processed, and extracting key labels in the sample business content;

and presenting a recommendation interface by taking the reason guide factors as case abstracts of the target business cases and the key tags.

In order to solve the foregoing technical problem, an embodiment of the present application further provides a recommendation device for a business case, including:

the acquisition module is used for acquiring the current service information of the current service case, wherein the current service information comprises at least one current label;

the extraction module is used for extracting a sample business case group with at least one current label in a data asset library based on the current label, wherein the sample business case group comprises at least one sample business case;

the analysis module is used for analyzing each sample business case in the sample business case group by applying NLP semantics and obtaining sample business information according to an analysis result;

the scoring module is used for establishing a scoring matrix according to the current business information of the current business case, the sample business information of the sample business case and a prestored evaluation score, and calculating the similar dimension of the current business information of the current business case and the sample business information of the sample business case in the scoring matrix based on a collaborative filtering algorithm;

the determining module is used for acquiring a sample business case with the highest similarity dimension with the current business case as a target business case according to the scoring matrix;

and the recommending module is used for acquiring the target service information of the target service case and recommending the target service case.

Further, the analysis module includes:

the analysis submodule analyzes the language structure in each sample service case in the sample service case group by applying NLP semantics to obtain sample service information, wherein the sample service information comprises sample service content and sample service nodes;

and the sequencing submodule sequences the sample service contents according to the sample service nodes.

In order to solve the above technical problem, an embodiment of the present application further provides a computer device, which adopts the following technical solutions: comprising a memory having computer readable instructions stored therein and a processor implementing the steps of the method for recommending a business case as described above when executing the computer readable instructions.

In order to solve the above technical problem, an embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solutions: the computer readable storage medium has stored thereon computer readable instructions which, when executed by a processor, implement the steps of the method of recommending a business case as described above.

Compared with the prior art, the method comprises the steps of acquiring current business information of a current business case, wherein the current business information comprises at least one current label, extracting a sample business case group with the at least one current label in a data asset library based on the current label, wherein the sample business case group comprises at least one sample business case, analyzing each sample business case in the sample business case group by applying NLP semantics, obtaining sample business information according to an analysis result, establishing a scoring matrix according to the current business information of the current business case, the sample business information of the sample business case and a prestored evaluation score, and calculating the similar dimension of the current business information of the current business case and the sample business information of the sample business case in the scoring matrix based on a collaborative filtering algorithm, and acquiring a sample business case with the highest similarity dimension with the current business case as a target business case according to the scoring matrix, acquiring target business information of the target business case, and recommending the target business case. The business case with high similarity is accurately recommended, the worker can work for reference according to the recommended business case, the worker can work quickly and accurately, the working efficiency is increased, and the intelligent level of work development is improved.

Drawings

In order to more clearly illustrate the solution of the present application, the drawings needed for describing the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and that other drawings can be obtained by those skilled in the art without inventive effort.

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

FIG. 2 is a first flowchart of an embodiment of a method for recommending business cases;

FIG. 3 is a flow chart two of one embodiment of a method for recommending business cases;

FIG. 4 is a schematic diagram of an embodiment of a recommendation device for business cases;

FIG. 5 is a schematic block diagram of one embodiment of a computer device according to the present application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "including" and "having," and any variations thereof, in the description and claims of this application and the description of the above figures are intended to cover non-exclusive inclusions. The terms "first," "second," and the like in the description and claims of this application or in the above-described drawings are used for distinguishing between different objects and not for describing a particular order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The

terminal devices

101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.

The

terminal devices

101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture experts Group Audio Layer III, mpeg compression standard Audio Layer 3), MP4 players (Moving Picture experts Group Audio Layer IV, mpeg compression standard Audio Layer 4), laptop portable computers, desktop computers, and the like.

The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the

terminal devices

101, 102, 103.

It should be noted that the information retrieval method based on the voice semantics provided by the embodiment of the present application is generally executed by a server, and accordingly, the information retrieval device based on the voice semantics is generally disposed in the server.

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 continuing reference to FIGS. 2 and 3, a flow diagram of one embodiment of a business case recommendation method according to the present application is shown. The business case recommendation method comprises the following steps:

step S1, obtaining current service information of a current service case, wherein the current service information comprises at least one current label;

the current business case is a business case which is to be processed by business personnel, the current label is a label generated by analyzing the current business case through NLP semantics, and the business personnel can also mark the label of the current business case according to the current business case situation, wherein the NLP semantics analysis is a machine learning algorithm based on a deep neural network, and combines natural language processing technologies such as Chinese participle, part of speech tagging, named entity recognition, syntactic analysis, semantic analysis and the like to carry out semantic understanding, so that intelligent recognition is realized on the basis of semantic understanding. In this embodiment, the current service information may include one current tag or a plurality of current tags, where the plurality of current tags are tags of different types.

Step S2, based on the current label, extracting a sample business case group with at least one current label in a data asset library, wherein the sample business case group comprises at least one sample business case;

in the application, a sample business case with the same label as the current business information is searched through the current label, and the first screening of the sample business case in a Data Asset library is completed, wherein the Data Asset library (Data Asset) refers to Data resources owned or controlled by an enterprise and capable of generating value for the enterprise, such as a client list, a transaction record, a business case processing record, a business round-trip time record and the like. Certainly, because the current business information may include a plurality of current tags, when the current business information includes a plurality of current tags, in the extraction of the data asset library, the sample business case needs to include at least one current tag and also can include a plurality of current tags at the same time, and the sample business case including at least one current tag is extracted to form a sample business case group, it needs to be noted that the sample business case group is all sample business cases including the current tags, the sample business case group includes one or more sample business cases, and the following content is to analyze each sample business case and calculate the similar dimension to the current business case.

Step S3, each sample business case in the sample business case group is analyzed by applying NLP semantics, and sample business information is obtained according to the analysis result;

analyzing each deep neural network machine learning algorithm in the sample service case group by applying NLP semantic analysis, performing semantic understanding by combining natural language processing technologies such as Chinese word segmentation, part of speech tagging, named entity recognition, syntactic analysis, semantic analysis and the like, further analyzing the sample service cases and obtaining sample service information, wherein the sample service information comprises key labels, sample service nodes and sample service contents, the key labels are key labels of key words and coordinates for embodying the sample service cases, and are convenient for business personnel to quickly obtain corresponding sample service cases through key word searching, the sample service nodes are arranged for classifying and sequencing the sample service contents, wherein each sample service node is associated with corresponding sample service contents, and in the embodiment, the sample service nodes are time nodes, namely time nodes for generating the sample service contents, the sample service content is a sample service event occurring in each sample service node in the sample service case and sample service processing content.

The method comprises the following steps of analyzing each sample business case in the sample business case group by applying NLP semantics, and obtaining sample business information according to an analysis result, wherein the steps comprise:

step S3 includes step S31, analyzing the language structure of each sample service case in the sample service case group by NLP semantics, and obtaining sample service information, where the sample service information includes sample service content and sample service nodes;

in a sample business case, description of the sample business case generally involves lexical, syntactic, pragmatic, contextual, natural language processing, i.e., the language structure is a semantic composition structure of lexical, syntactic, pragmatic, contextual, natural language processing. Specifically, the lexical analysis includes two aspects of morphological analysis and lexical analysis, wherein the morphological analysis is mainly expressed in analyzing prefixes, suffixes and the like of words, and the lexical analysis is expressed in controlling the whole lexical system, so that the characteristics of the input sample business case can be accurately analyzed, and the searching process is finally accurately completed; the syntactic analysis is to analyze the vocabulary phrases of the natural language of the input sample business case, and aims to identify the syntactic structure of a sentence so as to realize the process of automatic syntactic analysis; the pragmatic analysis is compared with semantic analysis, and the analysis of context, language background, context and the like is added, namely additional information such as image, interpersonal relationship and the like is extracted from the structure of an article, the pragmatic analysis is a higher-level linguistic analysis, and the content in a sentence is associated with the details in real life, so that a dynamic ideographic structure is formed; the context analysis mainly refers to a technology for analyzing a large number of 'gaps' outside the original query language so as to more accurately explain the language to be queried, and the 'gaps' comprise general knowledge, knowledge in a specific field and the like; the AI-driven engine can generate descriptions from the collected data, a software engine that creates seamless interactions between people and technology by following rules that convert results in the data into prose, natural language generation receiving the semantics of the structured representation to output grammatically-compliant, fluent, natural language text consistent with the input semantics.

Specifically, the technical system of natural language processing application mainly includes natural language processing at a word level, natural language processing at a syntax level, and natural language processing at a chapter level. The word level analysis mainly comprises Chinese word segmentation, named entity identification, part of speech tagging, synonym word segmentation, word vectors and the like; the analysis of the syntax level mainly comprises dependency grammar analysis, word position analysis, semantic normalization, text error correction and the like; the analysis of the chapter level mainly comprises tag extraction, document similarity analysis, topic model analysis, document classification, clustering and the like.

The Chinese word segmentation is a process that a computer automatically segments a Chinese character sequence into words conforming to human semantic understanding according to a semantic model, and the word segmentation is a process of recombining continuous character sequences into word sequences according to certain specifications.

Named entity recognition refers to a technology for automatically recognizing entities with specific meanings, and is an important basis for applications such as information extraction, knowledge mapping, question-answering system, syntactic analysis, search engine, machine translation and the like.

Part-of-speech tagging refers to a procedure for tagging each word in the segmentation result with a correct part-of-speech. Specifically, the process of determining whether each word is a noun, a verb, an adjective or other part-of-speech, the part-of-speech tagging based on maximum entropy, the output part-of-speech based on statistical maximum probability, the part-of-speech tagging based on hidden markov models.

Synonym analysis, because of cultural differences in different regions, the problem of inconsistent description of input query words is likely to occur, a processing system needs to perform synonym, error correction and normalization processing on input of a user, and a synonym algorithm comprises a dictionary, encyclopedic entries, meta search data and context correlation mining.

The word vector technology is to convert words into dense vectors, the word vectors corresponding to similar words are similar, and in natural language processing application, the word vectors are input as features of a deep learning model. The effect of the final model is therefore largely dependent on the effect of the word vectors, which in general are represented in two ways: one-hot means that only one dimension in a vector has a value of 1, and the other dimensions are 0, and this dimension represents a current word, distributed means that a word is converted into a distributed representation, which is also called a word vector, and distributed means that a word is represented as a dense vector of a fixed length.

The dependency grammar explains the syntactic structure by analyzing the dependency relationship before the components in the language unit, and the core predicate verb in the sentence is claimed to be the central component which dominates other components, but is not dominated by any other components, and all dominated components depend on a dominator in a certain relationship.

The word position analysis shows that the contribution degrees of words at different positions in an article to the article semantics are different, the probability that the words appearing at the head and the tail of the article become subject words and keywords is greater than that of the words appearing in the text, the positions of the words in the article are modeled, different weights are given to different positions, and therefore the article can be better represented in a vectorization mode.

Semantic normalization generally refers to identifying words or phrases from articles with the same meaning, the main task of which is coreference resolution. Coreference resolution is a core problem in natural language processing, mainly using a modeling system of common information extraction, microsoft's academic search engine will store some author's archives, and some of the information may be extracted according to coreference objects.

The text error correction task refers to automatically recognizing and correcting errors occurring during use of natural language. The text error correction task mainly comprises two subtasks, namely error identification and error correction. The task of error recognition is to indicate the position of the sentence where the error occurs, and error correction is to automatically correct on the basis of recognition, and the main difficulty of chinese correction compared to english correction lies in the linguistic characteristics of chinese: word boundaries in chinese and a large character set in chinese. The error types of the two languages are also different due to the linguistic characteristics of Chinese. The english modification operations include insertion, deletion, replacement and movement (movement means two letter exchange order, etc.), while for chinese, because each chinese kanji can be independently formed into a word, the errors of insertion, deletion and movement are only used as grammatical errors, and chinese input error correction is mainly focused on replacement errors.

The key labels are usually several words or phrases, and serve as a brief summary of the main contents of the sample business case. The key label is an important way for business personnel to quickly know document content and grasp subject, and the key label generally has the characteristics of readability, relevance, coverage and the like. Readability means that it should be meaningful as a word or phrase in itself; relevance means that the label must be closely related to the subject and content of the business case; coverage means that key tags can better cover the content of a document, and cannot be concentrated in a certain sentence.

The text similarity has different connotations due to different application scenes, so that no uniform definition exists. From the perspective of information theory, the similarity is related to the commonalities and the differences between texts, and the greater the commonalities, the smaller the differences, the higher the similarity; the smaller the commonality is, the larger the difference is, the lower the similarity is; the case of maximum similarity is where the text is identical. Similarity calculation generally refers to calculating the distance between features of objects, and if the distance is small, the similarity is large; if the distance is large, the similarity is small. Methods for similarity calculation can be divided into four major categories: string-based methods, corpus-based methods, knowledge-based methods, and other methods. The method based on the character strings is based on the matching degree of the character strings and takes the co-occurrence and the repetition degree of the character strings as the measurement standard of the similarity; the method based on the corpus is to calculate the similarity of texts by using information acquired from the corpus; knowledge-based methods refer to computing similarity of text using a knowledge base with a canonical organizational hierarchy.

The topic analysis model is a technology for counting and clustering implicit semantic structures of documents in an unsupervised learning mode so as to mine the semantic structures included in texts.

The text classification is according to the document classification system of the particular trade, the computer reads the content of the file automatically and belongs to under the corresponding technological system of category, wherein, typical processing procedure can be divided into training and operation two kinds. The computer reads documents of various categories in advance and extracts features, supervised learning training is completed, and the content of a new document is identified and classified in an operation stage.

Text clustering is mainly based on the well-known clustering assumption: the similarity of the similar documents is larger, the similarity of the different documents is smaller, the method is used as an unsupervised machine learning method, the clustering does not need a training process and manual labeling of the categories of the documents in advance, so that the method has certain flexibility and higher automatic processing capacity, and the text clustering method mainly comprises a clustering algorithm based on division, a clustering algorithm based on hierarchy and a clustering algorithm based on density.

After analyzing the language structure in the sample service case by NLP semantics, the analyzed sample service content and the sample service node are obtained, wherein the sample service content analyzed according to the specific analysis technology comprises a result key label, a case node of major concern, node time, node service content operated at the node time, customer attributes, a service processing result and reason guidance of success of the result. The key label information comprises the industry type of the client, the name of a commission agent lawyer, the name of a service processing person, the name of a service processing team, the related amount value, the type of overdue loan, the number of overdue days and the like; the key case-of-interest nodes comprise the stages of clearing and collecting, case division, litigation and asset recovery; the node time is the occurrence time of the key case-concerned node; the node service content operated at the node time specifically includes the customer attributes (XX industry customers, business owner lost clients stop production, asset amount XX, loan overdue days, asset type owned by the customer at present) when the clearing and accepting process starts; in the case stage, the node service content includes information according to the home principle, for example: distributing the sample business case to XX team XX people for completion, wherein an XX business manager has processing experience of XX years, and the recovery rate of the sample business case processed by the XX business manager in the optional period is what; in the litigation stage, the node business content comprises an entrusted XX attorney XX name for litigation; and recovering the assets, wherein the recovery result (the amount of the recovered assets and the overall recovery rate) is included in the node service content.

Step S32: and sequencing the sample service contents according to the sample service nodes.

The sample service content comprises key labels, key concerned case nodes, node time, node service content operated at the node time, customer attributes, service processing results and reason guidance of successful results, and specifically, sample service information is refined to facilitate similar dimension scoring in subsequent steps and increase scoring accuracy. The method and the device also process the arrangement of the sample service nodes and the sample service content, and arrange the sample service nodes according to the development time sequence of the sample service nodes, so that service personnel can conveniently check data, and the processing efficiency is improved.

Step S4, establishing a scoring matrix according to the current service information of the current service case, the sample service information of the sample service case and a pre-stored evaluation score, and calculating the similar dimension of the current service information of the current service case and the sample service information of the sample service case in the scoring matrix based on a collaborative filtering algorithm;

the method comprises the steps of classifying current business information of a current business case and sample business information of a sample business case and matching similar dimensions by applying a collaborative filtering algorithm, wherein the collaborative filtering algorithm is mainly used for discovering the development trend of the sample business case by mining historical behavior data of business personnel, classifying the sample business case in groups based on the development trends of different sample business cases and recommending target business cases with similar current labels.

Specifically, step S4 includes:

step S41: acquiring the pre-stored evaluation score and the current service node of the current service case, and acquiring the pre-stored evaluation score and the sample service node of the sample service case;

in the method, based on the application of a collaborative filtering algorithm, classification and similar dimension matching are carried out on the current service information of the current service case and the sample service information of the sample service case, and a collection case (target service case) adapted to the current service case is recommended according to a preset condition met by the similar dimension, wherein the preset condition is that the similar dimension meets a preset numerical value or the sample service case with the highest similar dimension is established, the preset numerical value is established according to the requirements of service personnel, and collection service case records and collection result evaluation, namely scoring, of key case nodes are achieved. After the sequencing, the gradient generated by the node time is compared with the similar dimension of the node service content in the node time, and then the current service information of the current service case is compared with the sample service information of the sample service case, and the similar dimension is obtained. The node business content specifically comprises the client attributes (XX industry clients, enterprise owner offline clients stop production, asset amount XX, loan overdue days and asset types owned by clients) when the clearing and accepting processing is started; in the case division stage, the node business content comprises that the case is distributed to XX team XX persons according to the home principle to be completed, wherein an XX business manager has XX-year processing experience, and the recovery rate of the case processed by the XX business manager in the leave period is what; in the litigation stage, the node business content comprises an entrusted XX attorney XX name for litigation; and recovering the assets, wherein the recovery result (the amount of the recovered assets and the overall recovery rate) is included in the node service content.

It should be noted that the pre-stored evaluation score is obtained by a service person according to the score of each service node and the score is recorded into the service case storage system when the service case information is recorded at the beginning, and is associated with the service case, so that when the current service case or the sample service case is obtained, the pre-stored evaluation score of the current service case and the pre-stored evaluation score of the sample service case can be extracted from the information of the current service case and the information of the sample service case, and the pre-stored evaluation score is the total evaluation score of the service case.

Step S42: splitting the pre-stored evaluation score of the current service case according to the current service node, acquiring the node score of at least one current service case, splitting the pre-stored evaluation score of the sample service case according to the sample service node, and acquiring the node score of the sample service case;

in order to improve the accuracy of the calculation result of the similar dimension, the pre-stored evaluation score is split according to each service node, so that the node score corresponding to each service node is obtained, and the node score is calculated to obtain the similar dimension. Specifically, splitting the pre-stored evaluation score of the current business case according to the current business node, obtaining the node score of at least one current business case, splitting the pre-stored evaluation score of the sample business case according to the sample business node, and obtaining the node score of at least one sample business case.

And filling node scores corresponding to the current service nodes in a table by taking the row head as the current service node and the column head as the name of the current service case, and establishing a current scoring submatrix, wherein the current service node is associated with the current service content.

In order to facilitate classification, the compared similar dimensions are scored, the head of a row is taken as a service node (the service nodes are sequentially sequenced according to node time), the head of a column is taken as a sample service case name (the current service case name), node scores of the service nodes are filled in the cells, the node scores in the cells are combined according to the sample service case name, and the similar dimensions of the current service case are calculated. The scoring matrix intuitively displays the similar dimensions between the plurality of sample business cases and the current business case, and the sample business cases with high similar dimensions are conveniently recommended to serve as reference bases for implementation of the current business case. Specifically, the head of a row is taken as the sample service node, the head of a column is taken as the name of the sample service case, and the node scores corresponding to the sample service node are filled in a table, namely a sample evaluation molecular matrix is established, wherein the sample service node is associated with the content of the sample service; and filling node scores corresponding to the current service nodes in a table by taking the row head as the current service node and the column head as the name of the current service case, namely establishing a current scoring submatrix, wherein the current service node is associated with the current service content.

Step S43: and calculating the similar dimensionality of the current scoring submatrix and the sample scoring submatrix based on a collaborative filtering algorithm.

Specifically, the product of the sum of the node scores of the current business case and the sum of the node scores of the sample business case is used as a dimension numerator, and the product of the sum of the node scores of the current business case squared and the sum of the node scores of the sample business case squared is used as a dimension denominator; and dividing the dimension numerator by the dimension denominator to obtain a value as the similar dimension of the current business case and the sample business case.

The calculation of the similar dimension includes a cosine algorithm, a modified cosine algorithm, and a pearson algorithm, and in this embodiment, the cosine algorithm is taken as an example, and the formula related to the cosine algorithm is as follows:

note: r _ u represents the total score of the current service case (i.e. a column of score data in the score matrix), r _ v represents the total score of the sample service case, i represents the service node, and Σ _i r _u，i r _v，i Represents the score of the current service case on service node 1 multiplied by the score of the sample service case on service node 1 plus the score of the current service case on service node 2 and so on added to the sum of the score of the last service node of the current service case multiplied by the score of the sample service case on the last service node,

the square of the score of the current business case on the business node 1 is added with the square of the score of the current business case on the business node 2, and the sum is added to the square of the score of the last business node of the current business case, so that the numerical value obtained by adding the square of the score of the sample business case on the business node 1, the square of the score of the sample business case on the business node 2, and the sum is added to the square of the score of the last business node of the sample business case, so that the numerical value obtained by adding the square of the score of the sample business case to the square of the last business node of the sample business case is used as a square root, and the square root of the current business case is multiplied by the square root of the sample business case, so that the similar dimension between the current business case and the sample business case is obtained.

And step S5, obtaining the sample business case with the highest similarity dimension with the current business case as the target business case according to the scoring matrix.

Calculating similar dimensions of the sample business case and the current business case according to the current scoring submatrix and the sample scoring submatrix, associating the sample business cases with the corresponding similar dimensions, sequencing the similar dimensions according to the numerical value, and taking the sample business case with the highest numerical value of the similar dimensions as a target case, wherein in another embodiment, a clearing case (target business case) adapted to the current business case is recommended according to a preset condition met by the similar dimensions, the preset condition is that the similar dimensions meet the preset numerical value or the sample business case with the highest numerical value of the similar dimensions, and the preset numerical value is set according to the requirements of business personnel.

Step S6, obtaining the target service information of the target service case and recommending the target service case.

In this embodiment, the target business case is a sample business case with the highest similarity to the current business case, the target business information of the target business case is used as an example, when a business person develops the current business case, the target business information of the target business case is used for reference, a data asset database of the target business case is extracted, similar target business cases can be used for reference in the clearing and collecting processes of different business cases, clearing and collecting strategies and suggestions are provided, the disposal efficiency of poor assets and the work development of recycling performance can be promoted, and the business person can develop business more quickly.

Step S7 is further included after the step S6, the reason leading factors of each key attention case node of the target business case, which are successfully processed, are extracted, and key labels in the sample business content are extracted;

step S8: and presenting a recommendation interface by taking the reason guide factors as case abstracts of the target business cases and the key tags.

When NLP semantically analyzes the target service information of a target service case, the target service information is presented to a recommendation interface according to the acquired key labels, the specifically presented target service information comprises the key labels of the target service case, customer attributes, case abstracts of key concerned case nodes and node service contents, the time for manually knowing the service case is saved, the brief case abstract contents can be extracted for a front-line service worker to quickly know excellent case clearing and accepting paths, the typesetting of the target service information on the recommendation interface can be preset when the target service information is input into the service case, can also be set according to the watching habits of the service worker, and is preset to input time sequencing or case first letter sequencing or case starting time sequencing.

The reason guide factors are specifically the recommendation of the processing content of each node, the recommendation of lawyers, the recommendation of an acquirer, the recommendation of a processing flow, the recommendation of a performance-preferred service person and the like. By recommending the case division process of the business case with higher similar dimension to business personnel, the recommendation result of each stage of each business case can be conveniently and dynamically obtained.

The processing method for displaying the target business case more intuitively on the recommendation interface saves manual time, can extract the case abstract file of the brief target business case, provides a front-line business personnel to quickly know the clearing and collecting path of the excellent business case, and promotes the disposal efficiency and the collection performance of the bad assets.

In this embodiment, current business information of a current business case is acquired, where the current business information includes at least one current tag, a sample business case group having at least one current tag in a data asset library is extracted based on the current tag, the sample business case group includes at least one sample business case, each sample business case in the sample business case group is analyzed by applying NLP semantics, sample business information is obtained according to an analysis result, a scoring matrix is established according to the current business information of the current business case, the sample business information of the sample business case, and a pre-stored evaluation score, a similarity dimension between the current business information of the current business case and the sample business information of the sample business case in the scoring matrix is calculated based on a collaborative filtering algorithm, and a sample business case with the highest similarity dimension with the current business case is acquired according to the scoring matrix as a target business case And acquiring the target service information of the target service case and recommending the target service case. The business case with high similarity is accurately recommended, the worker can work for reference according to the recommended business case, the worker can work quickly and accurately, the working efficiency is increased, and the intelligent level of work development is improved.

It is emphasized that to further ensure the privacy and security of the above-mentioned service cases, the above-mentioned service cases may also be stored in nodes of a block chain.

The block chain referred by the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.

The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.

The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware associated with computer readable instructions, which can be stored in a computer readable storage medium, and when executed, the processes of the embodiments of the methods described above can be included. The storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless explicitly stated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

With further reference to fig. 4, as an implementation of the method shown in fig. 2 and fig. 3, the present application provides an embodiment of a recommendation apparatus for business cases, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2 and fig. 3, and the apparatus may be specifically applied to various electronic devices.

As shown in fig. 4, the recommendation apparatus 400 for business cases according to this embodiment includes: an acquisition module 401, an extraction module 402, an analysis module 403, a scoring module 404, a determination module 405, and a recommendation module 406. Wherein:

an obtaining module 401, configured to obtain current service information of a current service case, where the current service information includes at least one current tag;

an extracting module 402, configured to extract, based on the current tag, a sample business case group having at least one current tag in the data asset library, where the sample business case group includes at least one sample business case;

an analysis module 403, configured to analyze each sample service case in the sample service case group by using NLP semantics, and obtain sample service information according to an analysis result;

the scoring module 404 is used for establishing a scoring matrix according to the current service information of the current service case, the sample service information of the sample service case and a pre-stored evaluation score, and calculating the similar dimension of the current service information of the current service case and the sample service information of the sample service case in the scoring matrix based on a collaborative filtering algorithm;

the determining module 405 acquires a sample business case with the highest similarity dimension with the current business case as a target business case according to the scoring matrix;

and the recommending module 406 acquires the target service information of the target service case and recommends the target service case.

In some optional implementations of this embodiment, the analysis module 403 includes:

In this embodiment, the NLP semantic analysis is used to analyze the language structure in each sample service case in the sample service case group, so as to obtain the detailed content of the sample service information, increase the authenticity of the data, and sort the sample service contents according to the sample service nodes, so that the service case content is more concise and clear.

In some optional implementations of this embodiment, the scoring module 404 includes:

the acquisition subunit is used for acquiring the pre-stored evaluation score and the current service node of the current service case and acquiring the pre-stored evaluation score and the sample service node of the sample service case;

the splitting subunit splits the pre-stored evaluation scores of the current service case according to the current service nodes, acquires the node score of at least one current service case, and establishes a current scoring submatrix according to the node score of the current service case; splitting the pre-stored evaluation scores of the sample business case according to the sample business nodes, acquiring at least one node score of the sample business case, and establishing a sample evaluation molecular matrix according to the node scores of the sample business case;

and the calculation sub-module is used for calculating the similar dimensionality of the current scoring sub-matrix and the sample scoring sub-matrix based on a collaborative filtering algorithm.

In the embodiment, the node scores are simpler and clearer by establishing the scoring matrix, and the node scores of the current service cases in the scoring matrix and the node scores of the sample service cases are subjected to similar dimension calculation, so that the service cases with high similarity can be accurately recommended.

In some optional implementations of this embodiment, the establishing sub-module includes:

establishing a current subunit, taking a head of a row as the current service node, taking a head of a column as the name of the current service case, filling a node score corresponding to the current service node in a table, and establishing a current scoring submatrix, wherein the current service node is associated with the current service content;

establishing a sample subunit, wherein the step of establishing a sample evaluation molecular matrix according to the node scores of the sample business cases comprises the following steps: and filling node scores corresponding to the sample service nodes in a table by taking the head of a line as the sample service node and the head of a column as the name of the sample service case, and establishing a sample evaluation molecular matrix, wherein the sample service node is associated with the content of the sample service.

In the embodiment, the pre-stored evaluation scores are split according to the service nodes, the node score of each service node is obtained, a scoring matrix is further established, and the accuracy of subsequent similar dimension calculation is improved.

In some optional implementations of this embodiment, the calculation submodule includes:

a calculating subunit, using a product of the sum of the node scores of the current business case and the sum of the node scores of the sample business case as a dimension numerator, and using a product of the sum of the node scores of the current business case squared and the sum of the node scores of the sample business case squared as a dimension denominator;

defining a subunit, and taking a value obtained by dividing the dimension numerator by the dimension denominator as a similar dimension of the current business case and the sample business case.

In the embodiment, the similarity dimension is calculated according to the node score of the current business case and the node score of the sample business case, so that the business case with high similarity can be accurately recommended.

In some optional implementation manners of this embodiment, the recommendation apparatus 400 for a business case further includes:

the extraction subunit is used for extracting reason guide factors for successful processing of each key concern case node of the target business case and extracting key labels in the sample business content;

and the presentation subunit is used for presenting a recommendation interface by taking the reason guide factor as the case abstract of the target business case and the key label.

In the embodiment, the processing method of the target business case is more intuitively displayed on the recommendation interface by extracting the reason guide factor for the successful processing of each key case node of the target business case, so that the manual time is saved, the case abstract file of the brief target business case can be extracted, a front-line business worker can quickly know the clearing and collecting path of the excellent business case, and the disposal efficiency and the recovery performance of the poor assets are promoted.

In order to solve the technical problem, the embodiment of the application further provides computer equipment. Referring to fig. 5, fig. 5 is a block diagram of a basic structure of a computer device according to the present embodiment.

The computer device 4 comprises a memory 41, a processor 42, a network interface 43 communicatively connected to each other via a system bus. It is noted that only computer device 4 having components 41-43 is shown in FIG. 5, but it is understood that not all of the illustrated components are required to be implemented, and that more or fewer components may alternatively be implemented. As will be understood by those skilled in the art, the computer device is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.

The computer device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The computer equipment can carry out man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch panel or voice control equipment and the like.

The memory 41 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 6. In other embodiments, the memory 41 may also be an external storage device of the computer device 6, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, provided on the computer device 4. Of course, the memory 41 may also include both internal and external storage devices of the computer device 4. In this embodiment, the memory 41 is generally used for storing an operating system installed in the computer device 4 and various application software, such as computer readable instructions of a recommendation method for a business case. Further, the memory 41 may also be used to temporarily store various types of data that have been output or are to be output.

The processor 42 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is configured to execute computer readable instructions or process data stored in the memory 41, for example, execute computer readable instructions of the recommendation method of the business case.

The network interface 43 may comprise a wireless network interface or a wired network interface, and the network interface 43 is generally used for establishing a communication connection between the computer device 4 and other electronic devices.

The present application further provides another embodiment, which is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, which can be executed by at least one processor, so as to cause the at least one processor to execute the steps of the recommendation method for business cases as described above.

In this embodiment, current service information of a current service case is obtained, where the current service information includes at least one current tag, a sample service case group having at least one current tag in a data asset library is extracted based on the current tag, the sample service case group includes at least one sample service case, each sample service case in the sample service case group is analyzed by applying NLP semantics, sample service information is obtained according to an analysis result, a scoring matrix is established according to the current service information of the current service case, the sample service information of the sample service case, and a pre-stored evaluation score, a similarity dimension between the current service information of the current service case and the sample service information of the sample service case in the scoring matrix is calculated based on a collaborative filtering algorithm, and a sample service case with the highest similarity dimension with the current service case is obtained according to the scoring matrix and is taken as a target service case And acquiring the target service information of the target service case and recommending the target service case. The business case with high similarity is accurately recommended, the worker can work for reference according to the recommended business case, the worker can work quickly and accurately, the working efficiency is increased, and the intelligent level of work development is improved.

Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.

It is to be understood that the above-described embodiments are merely illustrative of some, but not restrictive, of the broad invention, and that the appended drawings illustrate preferred embodiments of the invention and do not limit the scope of the invention. This application is capable of embodiments in many different forms and is provided for the purpose of enabling a thorough understanding of the disclosure of the application. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to one skilled in the art that the present application may be practiced without modification or with equivalents of some of the features described in the foregoing embodiments. All equivalent structures made by using the contents of the specification and the drawings of the present application are directly or indirectly applied to other related technical fields and are within the protection scope of the present application.

Claims

1. A method for recommending a business case, said method comprising the steps of:

2. The method for recommending business cases according to claim 1, wherein said analyzing each of said sample business cases in said set of sample business cases by NLP semantics, and obtaining sample business information according to the analysis result comprises:

and sequencing the sample service contents according to the sample service nodes.

3. The method according to claim 2, wherein the scoring matrix comprises a current scoring submatrix and a sample scoring submatrix, the step of establishing the scoring matrix according to the current business information of the current business case, the sample business information of the sample business case and a pre-stored evaluation score, and the step of calculating the similar dimension of the current business information of the current business case and the sample business information of the sample business case in the scoring matrix based on a collaborative filtering algorithm comprises:

4. The method for recommending service cases as claimed in claim 3, wherein said step of establishing a current rating matrix according to the node scores of said current service case comprises:

5. The method for recommending business cases of claim 4, wherein said step of calculating the similar dimension of said current scoring submatrix and said sample scoring submatrix based on a collaborative filtering algorithm comprises:

6. The method for recommending a business case according to claim 2, further comprising, after said step of obtaining target business information of said target business case and recommending said target business case:

7. A recommendation device for business cases, comprising:

8. The business case recommendation device of claim 7, wherein said analysis module comprises:

the analysis submodule analyzes the language structure of each sample service case in the sample service case group by applying NLP semantics and obtains sample service information, wherein the sample service information comprises sample service contents and sample service nodes;

9. A computer device, characterized in that it comprises a memory in which computer readable instructions are stored and a processor which, when executing said computer readable instructions, carries out the steps of the method for recommending service cases according to any of claims 1 to 6.

10. A computer readable storage medium, characterized in that it has computer readable instructions stored thereon, which when executed by a processor, implement the steps of the method of recommendation of business cases according to any of claims 1 to 6.