CN116452212A

CN116452212A - Intelligent customer service commodity knowledge base information management method and system

Info

Publication number: CN116452212A
Application number: CN202310448746.8A
Authority: CN
Inventors: 王冠鸿
Original assignee: Shenzhen Fast Selling Polytron Technologies Inc
Current assignee: Shenzhen Fast Selling Polytron Technologies Inc
Priority date: 2023-04-24
Filing date: 2023-04-24
Publication date: 2023-07-18
Anticipated expiration: 2043-04-24
Also published as: CN116452212B

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an intelligent customer service commodity knowledge base information management method and system. The method comprises the following steps: acquiring a real-time commodity data set, and carrying out commodity data cleaning and classifying treatment on the real-time commodity data set to acquire a standard commodity data set; establishing a mapping relation of information questions and answers of the commodity knowledge base by using a standard commodity data set, so as to generate an intelligent customer service commodity knowledge base information question and answer model; receiving commodity problem information sent by a terminal, transmitting the commodity problem information to an intelligent customer service commodity knowledge base information question-answering model for intelligent commodity information reply processing, and generating commodity question-answering information data; and carrying out user question and answer information optimization analysis processing on the commodity question and answer information data to generate commodity question and answer information optimization analysis data, and storing and transmitting the commodity question and answer information optimization analysis data to the terminal. The invention can intelligently answer the commodity information questions and solve the commodity questions proposed by the user in shopping.

Description

Intelligent customer service commodity knowledge base information management method and system

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to an intelligent customer service commodity knowledge base information management method and system.

Background

With popularization of the Internet and popularization and application of artificial intelligence, intelligent customer service is also increasing. The intelligent customer service is an industry-oriented application developed on the basis of large-scale knowledge processing, is suitable for technical industries such as large-scale knowledge processing, natural language understanding, knowledge management, automatic question-answering systems, reasoning and the like, provides fine-granularity knowledge management technology for enterprises, and establishes a quick and effective technical means based on natural language for communication between the enterprises and massive users; meanwhile, statistical analysis information required by fine management can be provided for enterprises; at present, most intelligent customer service is based on the application of big data knowledge processing technology, namely, data required by the intelligent customer service are stored in a knowledge base in advance, and when the intelligent customer service works, the knowledge stored in the knowledge base is read at any time. The data is composed of a large number of question-answer pairs, questions and corresponding answers are matched into a group, the question-answer pairs come from communication between usual visitors and artificial customer service or other dialogue occasions, and then a knowledge base for intelligent customer service is formed by manual combing; however, the answer to the answer by the traditional intelligent customer service is very mechanized, the requirement of the customer cannot be met, the answer questions cannot reach the effect of user satisfaction when the user questions about commodity, the user experience is reduced by the aid of a mechanically rigid answer mode, a large amount of dialogue data is stored in a knowledge base, the intelligent customer service has huge data quantity to be calculated when searching the answer every time, the required time is longer and longer, the response time of the answer questions of the intelligent customer service is affected, and the waiting time of visitors is prolonged.

Disclosure of Invention

The invention provides an intelligent customer service commodity knowledge base information management method, which solves at least one of the technical problems.

In order to achieve the above purpose, a method for managing information in an intelligent customer service commodity knowledge base comprises the following steps:

step S1: acquiring a real-time commodity data set, and carrying out commodity data cleaning and classifying treatment on the real-time commodity data set so as to acquire a standard commodity data set;

step S2: establishing a mapping relation of information questions and answers of the commodity knowledge base by using a standard commodity data set, so as to generate an intelligent customer service commodity knowledge base information question and answer model;

step S3: receiving commodity problem information sent by a terminal and transmitting the commodity problem information to an intelligent customer service commodity knowledge base information question-answering model for intelligent commodity information reply processing, so that commodity question-answering information data are generated;

step S4: and carrying out user question and answer information optimization analysis processing on the commodity question and answer information data to generate commodity question and answer information optimization analysis data, and storing and transmitting the commodity question and answer information optimization analysis data to the terminal.

The embodiment provides an intelligent customer service commodity knowledge base information management method, which is used for acquiring a real-time commodity data set collected from the outside, preventing result errors caused by messy or data redundancy and other problems in commodity data cleaning and sorting processing, enabling a user to answer intelligently by constructing an intelligent customer service commodity knowledge base information question-answering model, enabling the intelligent answer to be optimally answered and fed back to answers satisfied by the user, so that the answer of the intelligent customer service to the answer is intelligent, the requirement of the user is met, the effect of satisfaction of the user is achieved when the user questions are in question, the intelligent emotional answer mode is improved, the user experience is improved, a large amount of dialogue data is stored in a knowledge base, the data of the knowledge base is optimally processed, the accurate data amount is obtained when the intelligent customer service searches the answer each time, the required time is shorter, the response time of the intelligent customer service answer to the question is shortened, and the waiting time of the user is shortened.

In one embodiment of the present specification, step S1 includes the steps of, wherein the commodity data washing classification process includes a commodity data blocking process, a commodity data washing process, a commodity data noise reduction process, and a commodity data classification process:

step S11: acquiring a real-time commodity data set;

step S12: and carrying out commodity data blocking processing on the real-time commodity data set so as to generate a block file commodity data set.

Step S13: carrying out commodity data cleaning treatment on the commodity data set of the block file, and removing redundant data in the real-time commodity data set so as to obtain a cleaned commodity data set;

step S14: carrying out commodity data noise reduction processing on the cleaning commodity data set by utilizing a commodity data set noise reduction formula so as to generate a noise reduction commodity data set;

step S15: and carrying out commodity data classification processing on the noise reduction commodity data set according to the same data field, thereby generating a standard commodity data set.

In the embodiment, the data cleaning and classifying process is performed on the real-time commodity data set, and the data volume of the real-time commodity data set is too huge, the real-time commodity data set is firstly subjected to the blocking process, so that the pressure of a system on data processing is reduced, a large amount of repeated, useless and missing data exists in the data set, the useless and missing data are deleted, the missing value data is supplemented by a mean value, noise data in the data are processed, the data operation is prevented from being influenced, and when the data is acquired, different data cannot be directly used in the subsequent steps, the data is classified in the same data field, so that the data operation speed can be increased in the subsequent steps, and the accurate result can be obtained.

In one embodiment of the present disclosure, the merchandise dataset noise reduction formula in step S14 is:

the table is the noise reduction index of the commodity data set, < + >>The last item of wash commodity data expressed as a wash commodity data set,/->Weight information expressed as each cleaning commodity data, < +.>Weight information expressed as cleaning commodity data set restoration noise +.>Weight information indicating abnormal noise of the washed merchandise data set,/->To represent the average value of the abnormal noise of the washed merchandise data set,/->An adjustment term expressed as noise reduction index for the abnormal commodity data set.

In this embodiment, each item of data of the washed commodity data set is subjected to noise reduction processing, the commodity data set noise reduction index is obtained by washing the commodity data set, noise reduction processing is performed, and weight information of each item of washed commodity data is utilizedWeight information of the cleaning commodity restoration noise is obtained>Utilize->Noise repairing is carried out on the cleaning commodity data set, so that the cleaning commodity data set is helped to be noise-reduced, and the weight information of abnormal noise points of the cleaning commodity data set is +.>Mean value of abnormal noise point of data set of cleaning commodity +.>Obtain->，/>Representing analysis data obtained by cleaning abnormal noise points of commodity data set, and reusing adjustment item of noise reduction index of common commodity data set +. >Noise reduction index for commodity data set>And adjusting to complete the commodity data set noise reduction formula.

In one embodiment of the present specification, step S2 includes the steps of:

step S21: acquiring a commodity knowledge base information question-answering model construction module;

step S22: extracting data characteristic key fields from the standard commodity data set so as to generate a commodity key field data set;

step S23: performing neural network learning processing on the commodity key field data set and integrating the commodity key field data set so as to generate a commodity knowledge information base;

step S24: carrying out model construction processing on the commodity knowledge base by utilizing a commodity knowledge base information question-answer model construction module to obtain a commodity knowledge base information question-answer model;

step S25: and carrying out question-answer information optimization processing on the commodity knowledge base information question-answer model by utilizing the intelligent commodity knowledge base information question-answer formula, so as to generate the intelligent customer service commodity knowledge base information question-answer model.

According to the embodiment, an intelligent customer service commodity knowledge base information question-answering model is built, after commodity questions are presented by a user, the intelligent customer service commodity knowledge base information question-answering model is finally transmitted to help the user solve the required questions, answers to the questions are fed back to the user, characteristic information of standard commodity data sets is collected in the intelligent customer service commodity knowledge base information question-answering model, the characteristic data sets of the standard commodity data sets are marked as answers through neural network learning, accordingly, the reversely-pushed questions are stored in the intelligent customer service commodity knowledge base information question-answering model, when similar questions are presented by the user, the intelligent customer service commodity knowledge base information question-answering model can feed back answers to the user, efficiency is improved, waiting time of the user is shortened, and the answered questions accurately meet customer requirements.

In one embodiment of the present disclosure, the intelligent commodity knowledge base information question-answering formula in step S25 is:

the table is the optimization index of the commodity knowledge base information question-answer model,/-for>Represented as last item of merchandise information data in the merchandise knowledge information base->Weight information expressed as commodity data set information in commodity knowledge base information, +.>Weight information expressed as commodity key field data set in commodity knowledge base->Weight information expressed as optimal adjustment of each item of commodity data set in commodity knowledge base information, +.>Represented as an average value of information anomaly data of the commodity data set in the commodity knowledge base information,and optimizing index adjustment items for the abnormal commodity knowledge base information question-answer model.

In the embodiment, the intelligent commodity knowledge base information question-answer formula formed by each item of commodity information in the commodity knowledge base is utilized to carry out question-answer information optimization processing on the commodity knowledge base information question-answer model, and commodity keys in the commodity knowledge base are usedWeight information for field datasetsWeight information for optimal adjustment of each item of commodity data set in commodity knowledge base information +.>Optimization information for constructing each data set in commodity knowledge base information>Weight information of commodity data set information in commodity knowledge base information through optimizing information >Adjusting and utilizing average value of commodity data set information abnormal data in commodity knowledge base information +.>Optimizing and adjusting indexes, and optimizing index adjustment items by utilizing abnormal commodity knowledge base information question-answer model>And carrying out abnormal adjustment on the commodity knowledge base information question-answer model optimization index.

In one embodiment of the present specification, step S3 includes the steps of:

step S31: receiving commodity problem information sent by a terminal;

step S32: carrying out intelligent examination judgment processing on the commodity problem information so as to generate examination passing problem information or examination failing problem information;

step S33: transmitting the information of the auditing failure problem back to the terminal and feeding back;

step S34: and transmitting the audited through the problem information to an intelligent customer service commodity knowledge base information question-answering model for carrying out commodity information intelligent answer processing, so as to generate commodity question-answering information data.

According to the method, the device and the system, commodity problem information sent by a user is received through the terminal, the commodity problem information is audited, the robot can be prevented from maliciously brushing the commodity problem information, the problem that the audit fails is fed back to the terminal for processing, the commodity problem information is audited, answers can be more targeted, and for the audited problem information, commodity information answers can be rapidly fed back to the user through the problem in an intelligent customer service commodity knowledge base information questioning and answering model, so that the user can be helped to answer questions.

In one embodiment of the present specification, step S32 includes the steps of:

step S321: presetting a commodity problem information auditing threshold;

step S322: carrying out data field overlapping comparison processing by utilizing commodity problem information and a commodity knowledge information base to obtain commodity problem auditing passing rate;

step S323: and comparing the commodity problem auditing passing rate with a commodity information auditing threshold, judging that the commodity problem auditing passes the commodity information auditing threshold when the commodity problem auditing passing rate reaches the commodity information auditing threshold, and judging that the commodity problem auditing fails to pass the commodity information auditing threshold when the commodity problem auditing passing rate does not reach the commodity information auditing threshold.

The preset commodity problem information auditing threshold value in this embodiment is used to compare commodity problem information, when the received commodity problem information is completely unrelated to the commodity, and the commodity information in the commodity knowledge information base is completely unrelated to the commodity problem information, the commodity information is considered as auditing failure problem information, filtering processing is performed on the auditing failure problem information, operating pressure when the system processes the commodity problem is reduced, and the auditing passing problem information can be better collected, and then characteristic information is collected by utilizing the auditing passing problem information to generate commodity problem characteristic information, and corresponding commodity information is generated by utilizing the commodity problem characteristic information, so that subsequent commodity problem information can be better replied.

In one embodiment of the present specification, step S4 includes the steps of:

step S41: historical information data collection is carried out on the commodity question-answer information data, so that historical commodity question-answer information data are generated;

step S42: carrying out question and answer information classification on the historical commodity question and answer information data to generate historical classified commodity question and answer information data, and carrying out question and answer classified information access times statistics processing on the historical classified commodity question and answer information data to generate classified commodity question and answer information access times data;

step S43: the commodity question-answer analysis formula is utilized to carry out the algorithm improvement on the classified commodity question-answer information access frequency data by using the classified question-answer information of the commodity as a data basis, so as to build a model, and a commodity question-answer information analysis data model is generated;

step S44: carrying out user question-answer information optimization analysis processing on the commodity question-answer information data through a commodity question-answer information analysis data model, so as to generate commodity question-answer information optimization analysis data;

step S45: and transmitting the commodity question-answer information optimization analysis data to the terminal and feeding back the terminal.

According to the method, historical commodity question and answer information is collected, the most commodity question information which is proposed by a user is observed, the commodity question information which is accessed by the user is counted, the problem information which is accessed by the user is analyzed, what commodity questions are required by the user, the problem is solved more pertinently according to the requirements of the commodity questions, or a commodity question and answer information analysis data model is built by using the problem information, the commodity question and answer information analysis data model is continuously updated according to the data basis of the proposal of the user on different types of commodity information questions, when new commodity question and answer information data is transmitted to the commodity question and answer information analysis data model, whether the commodity question and answer information is hot commodity question and answer information is judged, what the commodity class of data is, and the like are transmitted to a terminal, and the optimal analysis data is fed back to the terminal, so that the commodity question and answer information optimal analysis data can be fed back to the user, and the requirements of various users on the commodity information are met.

In one embodiment of the present specification, the merchandise question and answer analysis formula includes an access classification merchandise question information data formula, and step S43 includes the steps of:

preprocessing the access times data of the classified commodity question-answer information by using an access classified commodity question information data formula to obtain access classified question-answer information data;

the commodity question-answer analysis formula is utilized to carry out the modification of the access classification question-answer information data by the classification question-answer access information of the commodity through the algorithm, so as to establish a model, and a commodity question-answer information analysis data model is generated;

the access classification commodity problem information data formula is:

wherein ,data index indicating the problem of accessing classified goods, < + >>Last commodity question information expressed as commodity question-answer information data,/item>Weight information expressed as question information of each commodity,/item>Weight information expressed as commodity information data set corresponding to each commodity question information, ++>Characteristic information data represented as merchandise problem information data extraction, < > or->The weight information is abnormal problem weight information of commodity problem information;

the commodity question-answer analysis formula is as follows:

wherein ,expressed as commodity question and answer analysis data index, +. >Last merchandise question information represented as merchandise question-answer information data,/-merchandise question information>Data index indicating the problem of accessing classified goods, < + >>Weight information expressed as all commodity question information data in commodity question-answer information data, +.>Weight information expressed as commodity information data set corresponding to each commodity question information, ++>And the index adjustment item is expressed as abnormal commodity question and answer analysis data.

In the embodiment, the commodity question-answer analysis formula is utilized to carry out a model establishment process on the classified commodity question-answer information access frequency data, the commodity question-answer analysis formula comprises an access classified commodity question information data formula, the access classified commodity question-answer information access frequency data is preprocessed by the access classified commodity question information data formula, and access classified question-answer information data consisting of question data submitted by a user on a commodity, hot information data of the commodity, price data of the commodity and the like can be obtained, wherein the weight information of each commodity question information in the access classified commodity question information data formulaThe weight information of the commodity information data set corresponding to each commodity question information constitutes weight information of commodity question-answer information +.>Characteristic information data extracted by using commodity problem information data >Abnormal question weight information of product question information>Adjusting to obtain the data index of the problem information of the access classified commodity->Data index +.>Weight information of commodity information data set corresponding to each commodity question information>Index calculation is performed and data index adjustment items are analyzed by using abnormal commodity questions and answers>And (3) adjusting to obtain commodity question-answer analysis data indexes, and establishing a commodity question-answer information analysis data model by using the commodity question-answer analysis data indexes on the basis of commodity question-answer information data.

In one embodiment of the present disclosure, an intelligent customer service commodity knowledge base information management system includes:

at least one processor; the method comprises the steps of,

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

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the intelligent customer service good knowledge base information management method of any one of claims 1 to 9.

The embodiment provides an intelligent customer service commodity knowledge base information management system, which can realize any intelligent customer service commodity knowledge base information management method, is used for combining operation and data transmission media among various devices, receiving commodity information data sent by terminal equipment, performing data blocking, cleaning and noise reduction on the data, so as to obtain a data set to be processed, establishing a commodity question and answer information model by using the data set to be processed, processing commodity problem information transmitted by a terminal, analyzing and processing the obtained commodity question and answer information data, and mutually cooperating internal structures of the system, so that the intelligent customer service commodity knowledge base information management method is perfected.

In the embodiment of the application, a more convenient mode is brought to the intelligent customer service management commodity knowledge base, the pressure of system operation is reduced by carrying out data partitioning on obtained commodity information data, then data cleaning and data noise reduction are carried out on the data, errors caused by data disorder can be reduced, data classification is carried out on the data, the data in the same data field are classified into one type, for example, commodity price is one type, commodity information is one type, the system is enabled to process a data set more conveniently, an intelligent customer service commodity knowledge base information question-answering model is built on a data base of the processed commodity information data set, questions possibly asked by a user are reversely deduced through the commodity information data, and then the questions are transmitted into the model, so that answers to questions can be retrieved by a horse, the waiting time of the user is reduced, answers required by the user can be more attached, the answers to the questions and the answers are analyzed, the most types of commodity information searched by the user are collected, the questions searched by the user and the answers obtained by the user are analyzed, the analyzed data are obtained, the intelligent customer service is enabled to obtain accurate answers to the questions when the customer service is searched each time, and the answer to the questions is shortened. And the optimization analysis data are transmitted to the terminal and fed back to the terminal, so that the optimization analysis data of the commodity question-answer information can be fed back to the user, and the requirements of each user on commodity information are met.

Drawings

FIG. 1 is a schematic flow chart of steps of an intelligent customer service commodity knowledge base information management method according to the present invention;

FIG. 2 is a flowchart illustrating the detailed implementation of step S2 in FIG. 1;

FIG. 3 is a flowchart illustrating the detailed implementation of step S3 in FIG. 1;

FIG. 4 is a flowchart illustrating the detailed implementation of step S4 in FIG. 1;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

The embodiment of the application provides an intelligent customer service commodity knowledge base information management method and system. The intelligent customer service commodity knowledge base information management method comprises the following steps of, but not limited to, carrying the system: the intelligent terminal device, the internet of things platform, the service site, the network transmission device and the like can be configured to execute at least one device of the method provided by the embodiment of the application. The internet of things platform includes, but is not limited to: logistics information platform, online mall platform, intelligent library platform, etc. The intelligent terminal includes, but is not limited to: smart phones, PCs with built-in multimedia functions, etc.

The invention provides an intelligent customer service commodity knowledge base information management method, which comprises the following steps:

The embodiment provides an intelligent customer service commodity knowledge base information management method, which is used for acquiring a real-time commodity data set collected from the outside, preventing result errors caused by messy or data redundancy and other problems in commodity data cleaning and sorting processing, enabling a user to answer intelligently by constructing an intelligent customer service commodity knowledge base information question-answering model, enabling the intelligent answer to be optimally answered and fed back to answers satisfied by the user, so that the answer of the intelligent customer service to the answer is intelligent, the requirement of the user is met, the answer questions reach the effect of user satisfaction when the user questions are in question, the intelligent emotional answer mode is improved, the user experience is improved, a large amount of dialogue data is stored in a knowledge base, the data of the knowledge base is optimally processed, the required time is shorter and shorter when the intelligent customer service searches the answer each time, and the waiting time of the user is shortened.

In the embodiment of the present invention, as described with reference to fig. 1, the steps of the data processing method applied to the cloud platform in this example include:

in the embodiment of the invention, the commodity data set which is updated in real time and is obtained from the shopping mall system is obtained, and the commodity data set is subjected to data cleaning and classifying treatment, so that the time for the system to run the commodity data set is reduced, and the result error caused by data disorder is reduced, thereby generating the standard commodity data set.

in the embodiment of the invention, a standard commodity data set is collected to establish an information base of the commodity data set, and a problem possibly mentioned by a user is calculated by utilizing the commodity data set as a data base, for example, the problem possibly asking the commodity by the user is calculated through the commodity price, so that an intelligent customer service commodity knowledge base information question-answering model is obtained.

in the embodiment of the invention, commodity question information sent by a user through a terminal is received and is transmitted into an intelligent customer service commodity knowledge base information question-answering model, and the intelligent customer service commodity knowledge base information question-answering model gives answers and replies according to the question information with extremely high similarity in the model, so that commodity question-answering information data is generated.

In the embodiment of the invention, the commodity question-answer information is collected and analyzed, if the commodity price information proposed by the user is the most of the problems or the function information of the commodity, the question-answer information is analyzed and processed to generate commodity question-answer information optimizing analysis data, and the commodity question-answer information optimizing analysis data is backed up and stored to prevent data loss and transmitted to the intelligent collection or the PC for feedback to the client.

step S11: acquiring a real-time commodity data set;

In the embodiment of the invention, a commodity data set which is updated in real time and is obtained from a shopping mall system is obtained, the commodity data set is divided into scattered block file data, so that the data parallel operation speeds up the follow-up steps, redundant data in the data set is removed, missing values are filled, the data is reduced by using a noise reduction commodity data set noise reduction formula, the data is more close to real data, the disordered data set in the data set is classified, such as commodity price is one type, commodity detailed information is one type and the like, and the data set is more convenient to use when the data is distinguished, and the condition that commodity price data is seen in commodity detailed information is avoided, so that a standard commodity data set is generated.

the table is the noise reduction index of the commodity data set, < + >>The last item of wash commodity data expressed as a wash commodity data set,/->Weight information expressed as each cleaning commodity data, < +.>Weight information expressed as cleaning commodity data set restoration noise +.>Weight information indicating abnormal noise of the washed merchandise data set,/->To represent the average value of the abnormal noise of the washed merchandise data set,/- >An adjustment term expressed as noise reduction index for the abnormal commodity data set.

In this embodiment, each item of data of the washed commodity data set is subjected to noise reduction processing, the commodity data set noise reduction index is obtained by washing the commodity data set, noise reduction processing is performed, and weight information of each item of washed commodity data is utilizedWeight information of the cleaning commodity restoration noise is obtained>Utilize->Noise repairing is carried out on the cleaning commodity data set so as to help the number of cleaning commoditiesNoise reduction is carried out on the data set by cleaning weight information of abnormal noise points of the commodity data set>Mean value of abnormal noise point of data set of cleaning commodity +.>Obtain->，/>Representing analysis data obtained by cleaning abnormal noise points of commodity data set, and reusing adjustment item of noise reduction index of common commodity data set +.>Noise reduction index for commodity data set>And adjusting to complete the commodity data set noise reduction formula.

In one embodiment of the present specification, step S2 includes the steps of:

As an example of the present invention, referring to fig. 2, a detailed implementation step flow diagram of step S2 in fig. 1 is shown, where the content in this example includes:

in the embodiment of the invention, the construction module for acquiring the commodity knowledge base information question-answering model is used for transmitting commodity information in and deducing the commodity questions possibly asked, and is the basis for constructing the intelligent customer service commodity knowledge base information question-answering model.

in the embodiment of the invention, for example, the data characteristic key field extraction is carried out on the standard commodity data set in the logistics information system, the logistics price, the arrival place, the required time and the like of the standard commodity data set are extracted, and the commodity key field data set is obtained by gathering the data.

in the embodiment of the invention, the information such as the logistics price, the arrival place, the logistics route and the like in the logistics information management system is used as an input layer to carry out neural network learning processing, and the possible problem which is possibly mentioned by a user is predicted to be an output layer, so that a knowledge information base of logistics information and corresponding logistics information problems is generated.

in the embodiment of the invention, the commodity knowledge base information question-answer model construction module is utilized to construct a model of the commodity knowledge base, and a mapping relation between commodity knowledge base information and question-answer is established so as to ensure a one-to-one correspondence relation between the question-answer and knowledge base information.

In the embodiment of the invention, the accuracy of the commodity knowledge base information question-answer model is optimized by utilizing the intelligent commodity knowledge base information question-answer formula, the question-answer data in the commodity knowledge base information question-answer model is matched in an optimized manner, the requirement of a user is met, the speed of replying to the user is higher on the premise of improving the accuracy, the answer quality of the commodity knowledge base information question-answer model is higher, and the stability of the commodity knowledge base information question-answer model is improved, so that the intelligent customer service commodity knowledge base information question-answer model is generated.

In the embodiment, the intelligent commodity knowledge base information question-answer formula formed by each item of commodity information in the commodity knowledge base is utilized to carry out question-answer information optimization processing on the commodity knowledge base information question-answer model, and the weight information of the commodity key field data set in the commodity knowledge base is utilizedWeight information for optimal adjustment of each item of commodity data set in commodity knowledge base information +.>Optimization information for constructing each data set in commodity knowledge base information>Weight information of commodity data set information in commodity knowledge base information through optimizing information >Adjusting and utilizing average value of commodity data set information abnormal data in commodity knowledge base information +.>Optimizing and adjusting indexes, and optimizing index adjustment items by utilizing abnormal commodity knowledge base information question-answer model>And carrying out abnormal adjustment on the commodity knowledge base information question-answer model optimization index.

In one embodiment of the present specification, step S3 includes the steps of:

step S31: receiving commodity problem information sent by a terminal;

in the embodiment of the invention, the user sends the problem information which the terminal wants to know to the system, the terminal receives and sends the problem information to the system, and the system needs to recognize, classify, analyze and the like after receiving the commodity problem information of the terminal.

in the embodiment of the invention, the intelligent examination and judgment processing of the problems is carried out on the commodity problem information, so that the quality and the correctness of the problems can be effectively ensured, and the problem that the system generates wrong answers due to the mistakes or inappropriateness of the problems is avoided. For example, if the questions sent by the terminal include sensitive words, improper words, etc., the system may automatically check for failure and return corresponding error information, marking the error information as check failure question information, and marking the normal merchandise question information as check passing question information.

In the embodiment of the invention, through the feedback of the question information of the failed examination, the user can be informed that the questions of the user cannot be answered and the reasons are explained, which is helpful for the user to know the limitation and the use requirement of the system and is also helpful for the user to put out more effective and reasonable questions. The problem information and the reasons thereof of the auditing failure are displayed on the terminal, so that the user can better understand the system and have higher satisfaction.

In the embodiment of the invention, the problem information which passes the audit is transmitted to the information question-answering model of the intelligent customer service commodity knowledge base, which is a key step for realizing intelligent answer. The intelligent customer service commodity knowledge base information question-answering model generates a large amount of commodity question-answering information by utilizing the mapping relation of the standard commodity data set, and optimizes the information by using an intelligent customer service commodity knowledge base information question-answering formula. Thus, the model can quickly answer the user's question about the merchandise. After the problem information which passes the auditing is transmitted to the intelligent customer service commodity knowledge base information question-answering model for processing, the system can generate accurate commodity question-answering information data, thereby providing more valuable and faster service for users.

In this embodiment, the method for managing information in the knowledge base of intelligent customer service commodity according to the present invention includes:

after commodity problem information is presented by a user, downloading is carried out through a specific template, commodity problem information is uploaded to an intelligent customer service commodity knowledge base information management system through a terminal by using the template, the intelligent customer service commodity knowledge base information management system processes after receiving a file, and after receiving the commodity problem information of the terminal, the system needs to process such processes as identification, classification, analysis and the like so as to further generate an answer through a commodity knowledge base information question-answer model, thereby realizing intelligent question-answer; and after the problems are uploaded, the problems are audited, an audit page has information such as commodity names, submitters, auditors, audit states and the like, the information data are submitted and audited after being filled, an intelligent customer service commodity knowledge base information management system audits commodity problem information, the validity and accuracy of information returned by intelligent customer service are ensured, the problems which do not meet the requirements are screened out, the quality of answers is ensured, and therefore the customer satisfaction is improved.

In one embodiment of the present specification, step S32 includes the steps of:

step S321: presetting a commodity problem information auditing threshold;

In the embodiment of the invention, a commodity problem information auditing threshold value used for judging whether commodity problem information passes is preset, in the process of auditing problem information, whether the problem information can pass the auditing is determined by a certain set standard, the standard can be the length, grammar structure, semantic content and the like of the problem, the purpose of the threshold value is to ensure the quality of the problem, the information replied by intelligent customer service is more reliable and the satisfaction degree of a user is improved, and the preset commodity problem information auditing threshold value needs to be set according to actual requirements and is evaluated and adjusted regularly; the commodity problem auditing passing rate can be obtained through superposition comparison processing of commodity problem information and commodity knowledge information base data fields, and the passing rate reflects the correlation of the problem information in the commodity knowledge information base when the system audits the commodity problem information; comparing commodity problem auditing passing rate with commodity information auditing threshold value, wherein the process is used for evaluating the integrity and accuracy of the problem information to determine whether the problem information can be used for generating an answer, the threshold value is a preset auditing standard, the passing rate is obtained through matching with a knowledge base, and if the passing rate reaches the threshold value, the problem information accords with the standard; if the threshold is not reached, the problem information is not in compliance with the standard.

In one embodiment of the present specification, step S4 includes the steps of:

According to the method, historical commodity question and answer information is collected, the most commodity question information which is proposed by a user is observed, the commodity question information which is accessed by the user is counted, the problem information which is accessed by the user is analyzed to obtain what commodity questions are required by the user, the problem is solved more pertinently according to the requirements of the commodity questions, or a commodity question and answer information analysis data model is built by using the problem information, the commodity question and answer information analysis data model is continuously updated according to the data basis of the proposal of the user on the commodity information which is different, when new commodity question and answer information data is transmitted to the commodity question and answer information analysis data model, whether the commodity question and answer information is popular commodity question and answer information, what the commodity class of data is, and the like can be judged, the optimization analysis data are transmitted to a terminal, feedback information can be transmitted to the user after the terminal receives feedback information.

As an example of the present invention, referring to fig. 4, a flowchart illustrating a detailed implementation step of step S4 in fig. 1 is shown, where the content in this example includes:

in the embodiment of the invention, for example, the historical information data collection is carried out on the commodity question-answer information of the logistics marketplace management system to obtain the problem information which is proposed by the user to the logistics marketplace management system, and the collection of the historical commodity question-answer information data of the logistics marketplace management system can help to improve the accuracy and efficiency of the intelligent customer service system and provide references for the commodity question answers. By continuously accumulating historical data, the intelligent customer service system can become more flexible and accurate when answering questions.

in the embodiment of the invention, for example, in a logistics mall management system, historical commodity question-answer information data is classified into commodity price question-answer information classes, commodity destination information question-answer information classes and the like, and the access times of different classified information are counted to generate access times data of the classified question-answer information.

in the embodiment of the invention, the classified commodity question-answer information access frequency data obtained in the steps is processed, the commodity classified question-answer access information is used as a data basis, a model is built through a commodity question-answer analysis formula improved algorithm, and a commodity question-answer information analysis data model is generated.

in the embodiment of the invention, the commodity question and answer information analysis data model is used for analyzing commodity question and answer information data to optimize user question and answer information, wherein the analysis possibly comprises statistics, prediction and classification of the data, and commodity question and answer information optimization analysis data is generated by utilizing the analysis results so as to help optimize the availability and effectiveness of commodity question and answer information.

In the embodiment of the invention, the commodity question-answer information optimization analysis data obtained about the commodity question-answer information in the commodity mall management system is transmitted to the terminal, and the terminal is fed back, so that the terminal can acquire and use the information, and the user acquires the feedback information from the terminal to use.

In the embodiment of the invention, a schematic diagram of the intelligent customer service commodity knowledge base information management method of the invention for optimizing and analyzing commodity question-answer information is transmitted to a terminal, and in the embodiment, the method comprises the following steps:

after the commodity question-answer information optimizing analysis data is transmitted to the terminal equipment, the terminal equipment receives and stores the commodity question-answer information optimizing analysis data, a user can view the commodity question-answer information optimizing analysis data on the terminal equipment, can see information such as data date, name, data quantity, modification quantity and the like inside the commodity question-answer information optimizing analysis data, and can derive the information.

the access classification commodity problem information data formula is:

the commodity question-answer analysis formula is as follows:

wherein ,represented as commodityQuestion-answer analysis data index, ->Last merchandise question information represented as merchandise question-answer information data,/-merchandise question information>Data index indicating the problem of accessing classified goods, < + >>Weight information expressed as all commodity question information data in commodity question-answer information data, +.>Weight information expressed as commodity information data set corresponding to each commodity question information, ++ >And the index adjustment item is expressed as abnormal commodity question and answer analysis data.

In the embodiment, the commodity question-answer analysis formula is utilized to carry out a model establishment process on the classified commodity question-answer information access frequency data, the commodity question-answer analysis formula comprises an access classified commodity question information data formula, the access classified commodity question-answer information access frequency data is preprocessed by the access classified commodity question information data formula, and access classified question-answer information data consisting of question data submitted by a user on a commodity, hot information data of the commodity, price data of the commodity and the like can be obtained, wherein the weight information of each commodity question information in the access classified commodity question information data formulaThe weight information of the commodity information data set corresponding to each commodity question information constitutes weight information of commodity question-answer information +.>Characteristic information data extracted by using commodity problem information data>Abnormal question weight information of product question information>Adjusting to obtain the data index of the problem information of the access classified commodity->Data index +.>Weight information of commodity information data set corresponding to each commodity question information >Index calculation is performed and data index adjustment items are analyzed by using abnormal commodity questions and answers>And (3) adjusting to obtain commodity question-answer analysis data indexes, and establishing a commodity question-answer information analysis data model by using the commodity question-answer analysis data indexes on the basis of commodity question-answer information data.

In the embodiment of the invention, the data model is established by using the commodity question and answer information access frequency information as a data basis and the commodity question and answer analysis formula, firstly, the data is preprocessed by using the commodity question and answer analysis formula to access the commodity question information data formula, so as to obtain the access classified question and answer information data consisting of the problem data presented by the commodity, the hot information data of the commodity, the price data of the commodity and the like, and then, the commodity question and answer analysis formula is used for establishing the data model by using the information data, so that the commodity question and answer information analysis data model is generated, and when commodity question and answer information enters the model, the commodity question and answer information analysis data model processes the question and answer information, thereby helping the statistical analysis of the commodity question and answer information and improving the effectiveness of the question and answer information.

At least one processor; the method comprises the steps of,

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

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the intelligent customer service inventory knowledge base information management method.

The embodiment provides an intelligent customer service commodity knowledge base information management system, which can realize any intelligent customer service commodity knowledge base information management method, is used for combining operation and data transmission media among all devices, receiving commodity information data sent by terminal equipment, performing data blocking, cleaning and noise reduction on the data to obtain a data set to be processed, establishing a commodity question and answer information model to process commodity problem information transmitted by a terminal, analyzing and processing the obtained commodity question and answer information data, and enabling internal structures of the system to cooperate with each other so as to complete the intelligent customer service commodity knowledge base information management method.

In the embodiment of the invention, in the intelligent customer service commodity knowledge base information management system, a real-time commodity data set transmitted by an online shopping mall shopping system transmitted by a terminal is collected and subjected to block, cleaning, noise reduction and classification processing so that problems caused by redundancy and disorder of data can be avoided when the system processes the data, commodity price information and commodity parameter information transmitted by the online shopping mall shopping system are classified, the data sets can be directly utilized by the system, an intelligent customer service commodity knowledge base information question-answering model for answering commodity information is established, commodity information is continuously collected and updated by the intelligent customer service commodity knowledge base information question-answering model, and problem information possibly appearing by the commodity information is calculated, after the problem information is received, the commodity information can be fed back to a smart phone immediately and is fed back to a user, commodity question-answering information is analyzed, and the favorite commodity, the user wants to search things and the like can be detected by the analysis data, so that the demand of each user on the commodity is met, and the user experience is remarkably improved.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.

It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The intelligent customer service commodity knowledge base information management method is applied to an intelligent customer service commodity knowledge base information management system and is characterized by comprising the following steps of:

2. The intelligent customer service commodity knowledge base information management method according to claim 1, wherein step S1 comprises the steps of, wherein the commodity data cleaning and sorting process comprises a commodity data blocking process, a commodity data cleaning process, a commodity data noise reduction process and a commodity data sorting process:

step S11: acquiring a real-time commodity data set;

step S12: carrying out commodity data blocking processing on the real-time commodity data set so as to generate a block file commodity data set;

3. The intelligent customer service commodity knowledge base information management method according to claim 2, wherein the commodity data set noise reduction formula in step S14 is:

the table is the noise reduction index of the commodity data set, < + >>Represented as the last item of cleaning article data in the cleaning article data set,weight information expressed as each cleaning commodity data, < +.>Weight information expressed as cleaning commodity data set restoration noise +.>Weight information indicating abnormal noise of the washed merchandise data set,/->To represent the average value of the abnormal noise of the washed merchandise data set,/->An adjustment term expressed as noise reduction index for the abnormal commodity data set.

4. The intelligent customer service commodity knowledge base information management method according to claim 2, wherein the step S2 comprises the steps of:

5. The intelligent customer service commodity knowledge base information management method according to claim 4, wherein the intelligent commodity knowledge base information question-answering formula in step S25 is:

the table is the optimization index of the commodity knowledge base information question-answer model,/-for>Represented as last item of merchandise information data in the merchandise knowledge information base->Weight information expressed as commodity data set information in commodity knowledge base information, +.>Weight information expressed as commodity key field data set in commodity knowledge base->Weight information expressed as optimal adjustment of each item of commodity data set in commodity knowledge base information, +.>Represented as the average value of the information anomaly data of the commodity data set in the commodity knowledge base information,/and the like>And optimizing index adjustment items for the abnormal commodity knowledge base information question-answer model.

6. The intelligent customer service commodity knowledge base information management method according to claim 4, wherein the step S3 comprises the steps of:

step S31: receiving commodity problem information sent by a terminal;

7. The intelligent customer service commodity knowledge base information management method according to claim 6, wherein step S32 comprises the steps of:

step S321: presetting a commodity problem information auditing threshold;

8. The intelligent customer service commodity knowledge base information management method according to claim 6, wherein the step S4 comprises the steps of:

9. The intelligent customer service commodity knowledge base information management method according to claim 8, wherein the commodity question-answer analysis formula includes an access classification commodity question information data formula, and step S43 includes the steps of:

the access classification commodity problem information data formula is:

the commodity question-answer analysis formula is as follows:

wherein ,expressed as commodity question and answer analysis data index, +.>Last merchandise question information represented as merchandise question-answer information data,/-merchandise question information>Data index indicating the problem of accessing classified goods, < + >>Weight information expressed as corresponding problem information data in commodity knowledge base information,/for example >Weight information expressed as commodity information data set corresponding to each commodity question information, ++>And the index adjustment item is expressed as abnormal commodity question and answer analysis data.

10. An intelligent customer service commodity knowledge base information management system, which is characterized by comprising:

at least one processor; the method comprises the steps of,

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