Semantic-based intelligent logistics customer service processing method and system

Abstract

The invention discloses a semantic-based intelligent logistics customer service processing method and system, and relates to the technical field of data service value evaluation and optimization, and the method comprises the steps: collecting data obtained by scoring data service application effect indexes by different users and index data describing user features; based on a multi-layer perceptron model, predicting score conditions of different feature users for application effects of different data services; and a comprehensive evaluation index system is constructed, comprehensive scores of different users for different data services are obtained, and a data service type with the highest comprehensive score is provided for typical users. According to the method, a data service application effect is predicted for a user lacking application score data by using a multi-layer perceptron based on historical data, application effects brought to typical users by different types of data services are analyzed, targeted data services and improvement suggestions are provided for each industry user, benefits are helped to be improved, and the user experience is improved. And effective utilization of the data service is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of data service value assessment and optimization, and specifically to a semantic-based intelligent logistics customer service processing method and system. Background Art

[0002] With the development of big data technology, the application of data services has gradually become an important means to improve operational efficiency and optimize decision-making in various industries. Data services help users improve accuracy and efficiency in analysis, decision-making and execution by providing a large amount of high-quality data information. However, due to the diversity of users in data service application scenarios, evaluating the value of data services and providing users with the most appropriate service type is still a complex process. At present, common data service value assessment methods include comprehensive evaluation based on data acquisition costs, market prices and user feedback. Although these methods can provide a certain reference basis, there are still problems such as low prediction accuracy and insufficient personalization. Therefore, intelligent evaluation and optimization of data services based on artificial intelligence language models has become a research focus of academia and industry.

[0003] Although existing technologies can evaluate the value of data services through cost-benefit method or market method, they have obvious deficiencies in personalized evaluation and matching user needs. Traditional methods rely on historical data and it is difficult to effectively predict the effects of data services for new users or in new scenarios. Existing evaluation methods fail to fully consider the characteristics and needs of different users and it is difficult to provide customized services for different users. The comprehensive evaluation model between the cost and effect of data services is relatively simple and fails to build a systematic and globally optimized evaluation framework. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: the existing data service evaluation methods are insufficiently personalized, cannot effectively predict the data service effects of new users, lack a systematic comprehensive evaluation framework, and how to provide the optimal data service type for typical users.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a semantic-based intelligent logistics customer service processing method, comprising collecting data on different users' scoring of data service application effect indicators and indicator data that characterize user characteristics; based on a multi-layer perceptron model, predicting the scores of application effects of different data services for users with different characteristics; constructing a comprehensive evaluation index system to obtain the comprehensive scores of different users for different data services, and providing typical users with the data service type with the highest comprehensive score; setting a user interaction function to interact with the platform through natural language instructions, parsing natural language instructions and generating corresponding standardized input file instructions and visualization instructions for model results; setting a code generation function to generate a script for adjusting the model input file, detect logical defects in the file, and generate a test plan and visualization results.

[0007] As a preferred solution of the semantic-based intelligent logistics customer service processing method described in the present invention, the data for different users to score the data service application effect indicators includes issuing questionnaires, requiring typical industry users to score the application effects of existing data services, and the application effects include basic data application effects, conventional data application effects, and advanced data application effects.

[0008] As a preferred solution of the semantic-based intelligent logistics customer service processing method described in the present invention, the indicator data that characterizes user characteristics includes collecting indicators that characterize typical users, including basic information, transaction behavior, and benefit and efficiency related indicators.

[0009] As a preferred solution of the semantic-based intelligent logistics customer service processing method described in the present invention, the multi-layer perceptron model includes indicators that characterize user conditions as independent variables and application effect scores as dependent variables. The multi-layer perceptron MLP is used to predict the evaluation scores of users with different characteristics on the application effects of data services. The architecture of the neural network is designed according to the actual data situation, including the number of input layers, hidden layers, and output layers and the number of nodes in each layer.

[0010] As a preferred solution of the semantic-based intelligent logistics customer service processing method described in the present invention, the scoring situation includes applying the trained multi-layer perceptron model to the data set, inputting the values ​​of relevant indicators for typical users, and obtaining the application effect score of a single data service for the user.

[0011] As a preferred solution of the semantic-based intelligent logistics customer service processing method described in the present invention, the construction of a comprehensive evaluation index system includes evaluating the value of data assets from two aspects: data asset cost and actual application effect based on the hierarchical analysis method and the fuzzy evaluation model, constructing a data asset value evaluation index system, and evaluating the actual value of data assets.

[0012] As a preferred solution of the semantic-based intelligent logistics customer service processing method described in the present invention, the data service type with the highest comprehensive score provided to typical users includes obtaining the comprehensive value scores of different users for different data services, comparing the comprehensive value scores of different data services to users, and recommending the data service type with the highest comprehensive value score and level to typical users to help users optimize their decisions.

[0013] Another object of the present invention is to provide a data service value evaluation and decision optimization system based on a multi-layer perceptron, which can predict the scores of the application effects of different data services for users with different characteristics based on a multi-layer perceptron model, thereby solving the problem that current data service evaluation technology cannot effectively predict the data service effects of new users.

[0014] As a preferred solution of the data service value evaluation and decision optimization system based on multi-layer perceptron described in the present invention, it includes: a data collection module, a score prediction module, and a comprehensive evaluation module; the data collection module is used to collect data on the scores of different users for the data service application effect indicators and the indicator data that characterizes the user characteristics; the score prediction module is used to predict the scores of the application effects of different data services for users with different characteristics based on the multi-layer perceptron model; the comprehensive evaluation module is used to construct a comprehensive evaluation index system to obtain the comprehensive scores of different users for different data services, and provide typical users with the data service type with the highest comprehensive score.

[0015] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of a semantic-based intelligent logistics customer service processing method.

[0016] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a semantic-based intelligent logistics customer service processing method.

[0017] Beneficial effects of the present invention: The semantic-based intelligent logistics customer service processing method provided by the present invention uses a multi-layer perceptron based on historical data to predict the application effect of data services for users who lack application score data. On the basis of relatively mature research results, the actual operating characteristics of the enterprise and the nature of the data assets are considered, and the hierarchical analysis model and fuzzy value evaluation method are used to quantify the actual value of the data assets of the supplier enterprise; by analyzing the application effects of different categories of data services on typical users, combined with data service component analysis, the data value is graded, and targeted data services and improvement suggestions are provided to each industry user to help them increase their profits and promote the effective use of data services. The present invention achieves better results in terms of reliability, cost and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0019] Figure 1 An overall flow chart of a semantic-based intelligent logistics customer service processing method provided for the first embodiment of the present invention.

[0020] Figure 2 An MLP model structure diagram of a semantics-based intelligent logistics customer service processing method provided for the second embodiment of the present invention.

[0021] Figure 3 An overall flow chart of a data service value assessment and decision optimization system based on a multi-layer perceptron is provided for the third embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0023] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a semantic-based intelligent logistics customer service processing method, including:

[0024] S1: Collect data on different users’ scores on data service application effect indicators and indicator data that characterize user characteristics.

[0025] Furthermore, the data on different users' scores on data service application effect indicators include those obtained by distributing questionnaires, which require typical industry users to score the application effects of existing data services. The application effects include basic data application effects, general data application effects, and advanced data application effects.

[0026] It should be noted that the indicator data that characterizes user characteristics includes the collection of indicators that characterize typical users, including basic information, transaction behavior, and benefit and efficiency related indicators.

[0027] It should also be noted that basic data applications mainly include the collection, storage, cleaning and integration of raw data; conventional data applications include daily reports, descriptive statistical analysis, etc.; advanced data applications include the use of statistical models and machine learning algorithms for prediction, data mining, automated decision-making and intelligent recommendations through artificial intelligence models, etc.

[0028] S2: Based on the multi-layer perceptron model, predict the scores of application effects of different data services for users with different characteristics.

[0029] Furthermore, the multi-layer perceptron model includes indicators that characterize user conditions as independent variables and application effect scores as dependent variables. The multi-layer perceptron MLP is used to predict the evaluation scores of users with different characteristics on the application effects of data services. The neural network architecture is designed according to the actual data situation, including the number of input layers, hidden layers, and output layers, as well as the number of nodes in each layer.

[0030] It should be noted that the scoring situation includes applying the trained multi-layer perceptron model to the data set, inputting the values ​​of relevant indicators for typical users, and obtaining the application effect score of a single data service for the user.

[0031] It should also be noted that the specific process of using the multi-layer perceptron MLP to predict the evaluation scores of users with different characteristics on the application effects of data services includes data preprocessing: performing operations such as data cleaning and feature normalization, as well as improving the accuracy and efficiency of the model; neural network architecture design: determining the number of input layers, hidden layers, and output layers, the number of nodes in each layer, and the selection of activation functions, where the input layer is the collected enterprise features, and the output layer is the evaluation score of the application effects of data services. Since this is a regression prediction problem, the activation function is set to the ReLU function, that is, f(x) = max(0, x); setting the loss function: for regression prediction problems, the loss function is defined as the mean square error, expressed as:

[0032]

[0033] Where MSE is the average of the sum of squares of prediction errors of all users. The smaller the value, the better the prediction effect of the model. N is the total number of users or data service samples participating in the evaluation. i is the actual score given by the ith user to a data service, representing the user’s actual experience of the service. It is the score of the i-th user for the data service predicted by the multi-layer perceptron model; model training: input the processed data into the model, and continuously adjust the weights and biases through the back propagation algorithm to gradually reduce the model loss function; model evaluation: use the test set to evaluate the model, calculate the model's accuracy, precision, recall rate and other indicators, and judge the model's performance; model application: apply the trained model to a new data set, and for new typical enterprise users, input their feature values ​​to obtain the application effect of the data service for them.

[0034] It should also be noted that the data asset value is evaluated from the two aspects of data asset cost and actual application effect, and a data asset value evaluation index system is constructed to evaluate the actual value of data assets. The index layer in the data asset value evaluation system is divided into four layers, among which the target layer is the data asset value composition, the category layer is the data asset cost and data asset application, and the index layer is the data information statistical cost, strategic data analysis cost, data value application cost, statistical data quality, data analysis quality and data in-depth application effect.

[0035] S3: Build a comprehensive evaluation index system to obtain the comprehensive scores of different users for different data services, and provide typical users with the data service type with the highest comprehensive score.

[0036] Furthermore, the construction of a comprehensive evaluation index system includes evaluating the value of data assets from the two aspects of data asset cost and actual application effect based on the hierarchical analysis method and fuzzy evaluation model, building a data asset value evaluation index system, and evaluating the actual value of data assets.

[0037] It should be noted that providing typical users with the data service type with the highest comprehensive score includes obtaining the comprehensive value scores of different users for different data services, comparing the comprehensive value scores of different data services for users, and recommending the data service type with the highest comprehensive value score and level to typical users to help users optimize their decisions.

[0038] Example 2, reference Figure 2-Figure 3 , which is an embodiment of the present invention, provides a semantic-based intelligent logistics customer service processing method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0039] First, we collected data on different users’ scores on a series of data service application effect indicators and indicator data that characterize these users’ characteristics; by distributing questionnaires, we asked multiple typical industry users to score the application effects of existing data services, including basic data application effects, conventional data application effects, and advanced data application effects. Basic data applications mainly include the collection, storage, cleaning, and integration of original data; conventional data applications include daily reports, descriptive statistical analysis, etc.; advanced data applications include prediction using statistical models and machine learning algorithms, data mining, automated decision-making through artificial intelligence models, and intelligent recommendations; in addition, we collected indicators that characterize these shipping companies, including their basic information, transaction behavior, efficiency, and future prospects; specific indicators such as Figure 2-Figure 3 As shown in the figure, basic information mainly includes three aspects: basic attributes, financial status and social credit; basic attributes cover registration information, hardware equipment, management capabilities, etc.; financial status is subdivided into profitability, debt-to-asset ratio, operating capacity and financing status; social credit focuses on corporate credit ratings and credit records in corporate public information; these factors are mainly used to evaluate the basic operating conditions and credit risks of enterprises; trading behaviors involve market-oriented trading capabilities, trading conditions, contract performance, operation management and information disclosure; market-oriented trading capabilities include sales scale, user coverage, etc., while trading conditions evaluate indicators such as power sales and annual transaction volume; contract performance, operation management and information disclosure focus on corporate contract performance capabilities, information transparency, etc. These dimensions jointly reflect the trading capabilities and behavioral performance of enterprises; efficiency and benefit efficiency are composed of two parts: agent user analysis and participation in medium- and long-term transaction analysis. Agent user analysis is refined into the scale of agent users and the coverage of retail business, while participation in medium- and long-term transaction analysis focuses on transaction price volatility and transaction volume analysis. , combined with the overall benefit and efficiency to evaluate the performance of the enterprise in the trading market; the future outlook focuses on the outlook for spot market transactions and medium- and long-term transactions, which is divided into analysis of actual trading performance and participation in monthly settlement. The former includes spot price volatility and risk-return ratio, and the latter is an expected assessment of the future performance and pricing trends of the enterprise in medium- and long-term contracts, reflecting the future market opportunities and risks of the enterprise as a whole; based on these data, the multi-layer perceptron model is used to predict the scores of the application effects of different data services for users with different characteristics. The specific process includes: data preprocessing: including data cleaning, feature normalization and other operations, as well as improving the accuracy and efficiency of the model; neural network architecture design: determine the number of input layers, hidden layers, and output layers, the number of nodes in each layer, and the selection of activation functions, where the input layer is the collected enterprise features, and the output layer is the evaluation score of the application effect of the data service. Since this is a regression prediction problem, the activation function is set to the ReLU function, that is, f(x) = max(0,x), and the MLP model structure is as follows Figure 2As shown; Set the loss function: For regression prediction problems, define the loss function as the mean square error; Model training: Input the processed data into the model, and continuously adjust the weights and biases through the back-propagation algorithm to gradually reduce the model loss function; Model evaluation: Use the test set to evaluate the model, calculate the model's accuracy, precision, recall rate and other indicators, and judge the model's performance; Model application: Apply the trained model to a new data set, and for new typical enterprise users, input their feature values ​​to obtain the application effect of the data service for them. Build a comprehensive evaluation index system. Evaluate the value of data assets from the two aspects of data asset cost and actual application effect, and build a data asset value evaluation index system to evaluate the actual value of data assets; specific indicators such as Figure 2 As shown in the figure, the indicator layer in the data asset value assessment system is divided into four layers, among which the target layer is the data asset value composition, the category layer is the data asset cost and data asset application, and the indicator layer is the data information statistical cost, strategic data analysis cost, data value application cost, statistical data quality, data analysis quality and data deepening application effect; then, the hierarchical analysis method is used to calculate the weight of each indicator, so that experts are isolated from each other and independently complete the assignment and scoring of each indicator, summarize, analyze and calculate the scoring results, obtain the indicator judgment matrix, and calculate the weight of each indicator. The specific process includes the weight calculation and consistency test of the category layer and the indicator layer; in the data asset value assessment indicators, the target layer is represented by A, the category layer is represented by B1 and B2, and the indicator layer is represented by C1, C2, ... C6 in turn, w i Indicates the value of each indicator weight, λ min represents the maximum eigenvalue of the judgment matrix, CI represents the consistency index, RI represents the average random consistency index, CR represents the consistency ratio, and the requirements for consistency are as follows:

[0040]

[0041] When CR<0.1, it indicates that the judgment matrix has satisfactory consistency. Otherwise, the judgment matrix needs to be perfected and improved so that it can finally have satisfactory consistency.

[0042] Table 11-9-order matrix average random consistency index value

[0043] Order n Average random consistency index RI 1 0 2 0 3 0.58 4 0.9 5 1.12 6 1.24 7 1.32 8 1.41 9 1.45

[0044] The calculated values ​​of the weights of each indicator at the category level obtained using the hierarchical analysis method are shown in Table 2.

[0045] Table 2 Judgment matrix A weight vector and consistency test

[0046] A <![CDATA[B1]]> <![CDATA[B2]]> <![CDATA[w i ]]> <![CDATA[B1]]> 1 2 0.6667 <![CDATA[B2]]> 1 / 2 1 0.3333

[0047] λ min =2,CI=0,RI=0,CR=0<0.1, indicating that the judgment matrix has satisfactory consistency.

[0048] The calculated values ​​of indicator weights for each indicator layer are shown in Table 3.

[0049] Table 3 Weight vector and consistency test of judgment matrix B1

[0050] <![CDATA[B1]]> <![CDATA[C1]]> <![CDATA[C2]]> <![CDATA[C3]]> <![CDATA[w i ]]> <![CDATA[C1]]> 1 2 3 0.5278 <![CDATA[C2]]> 1 / 2 1 3 0.3325 <![CDATA[C3]]> 1 / 3 1 / 3 1 0.1369

[0051] λ min =3.0536,CI=0.0268,RI=0.52,CR=0.0516<0.1, indicating that the indicator layer judgment matrix has satisfactory consistency.

[0052] Table 4 Weight vector and consistency test of judgment matrix B2

[0053] <![CDATA[B2]]> <![CDATA[C1]]> <![CDATA[C2]]> <![CDATA[C3]]> <![CDATA[w i ]]> <![CDATA[C1]]> 1 2 4 0.5714 <![CDATA[C2]]> 1 / 2 1 2 0.2857 <![CDATA[C3]]> 1 / 4 1 / 2 1 0.1429

[0054] λ min =3,CI=0,RI=0.52,CR=0<0.1, indicating that the indicator layer judgment matrix has satisfactory consistency.

[0055] By synthesizing the above results, the synthetic weight of the indicator layer relative to the target layer is obtained. The numerical values ​​of the weights of each indicator are shown in Table 5.

[0056] Table 5 Synthesis weights of sub-criteria layers relative to target layers

[0057]

[0058] Carry out weight calculation and consistency test of sub-indicator layer; in this data asset value assessment indicator system, the sub-indicator layer is data statistical environment creation cost, data creation cost, data security maintenance cost, information archiving cost, data information destruction cost, data fault repair cost, data operation and maintenance cost, actual application cost, technical operation cost, data update timeliness rate, data information completeness rate, data information accuracy rate, statistical analysis tool diversity, analysis process convenience, analysis result rationality, data basic application effect, data conventional application effect and advanced application effect. The sub-indicator layers are represented by D1, D2, ... D 18 Indicates; Since the sub-indicator layer contains more elements, the judgment matrix C1 is taken as an example to obtain the indicator weight value of the sub-indicator layer C1.

[0059] Table 6 Weight vector and consistency test of judgment matrix C1

[0060] <![CDATA[C1]]> <![CDATA[D1]]> <![CDATA[D2]]> <![CDATA[D3]]> <![CDATA[w i ]]> <![CDATA[D1]]> 1 3 2 0.5396 <![CDATA[D2]]> 1 / 3 1 1 / 2 0.1634 <![CDATA[D3]]> 1 / 2 2 1 0.2970

[0061] λ min =3.0092,CI=0.0046,RI=0.52,CR=0.0088<0.1, the CRs of C2~C6 are 0.0053, 0.0279, 0.0176, and 0.0707, respectively, all of which are less than the standard value of 0.1, indicating that the judgment matrices of the sub-indicator layer all have satisfactory consistency; the synthetic weights of the sub-indicator layer relative to the target layer are obtained by synthesizing the above calculated values, and the specific numerical values ​​of the weights of each indicator are shown in Table 7.

[0062] Table 7 Synthetic weights of sub-index layers relative to the target layer

[0063]

[0064]

[0065] After obtaining the weights of each indicator, combined with the evaluation levels of each indicator that actually need to be determined in the indicator system, the evaluation matrix and fuzzy evaluation model are constructed, and finally the scores of different enterprises for the value of data assets are calculated. The construction process specifically includes determining the main prime set U, which is:

[0066] U=(C1, C2, C3, C4, C5, C6)

[0067] Determine the sub-element evaluation indicator set and find its weight from the calculation table; specifically:

[0068] C1=(D1, D2, D3), C2=(D4, D5, D6),..., C6=(D 16 , D 17 , D 18 )

[0069] The weight is expressed as:

[0070] W1=(w1,w2,w3), W2=(w4,w5,w6),..., W6=(w 16 , w 17 , w 18 )

[0071] Determine the evaluation level set; according to the basic order of data service value from high to low, use the arithmetic scoring method to subdivide the data service value into three levels V = {v1, v2, v3): the high-level data asset value v1 is assigned a value of 65-100, the medium-level data service value v2 is assigned a value of 30-65, and the low-level data service value v3 is assigned a value of 0-30. Therefore, the specific data service value evaluation set is G = (100, 65, 30).

[0072] Construct a data asset value assessment matrix; in order to accurately study the membership of each evaluation indicator, invite multiple experts to conduct subjective judgments on subjective evaluation indicators; since the characteristics of indicators at the same level are quite different, the methods for determining their membership are also different; through the analysis of the above-constructed indicator system, select the trapezoidal distribution function to represent the cost of the data assets of the supplier enterprise, the benefits obtained by the application, and other indicators, among which the calculation method of the cost-type membership function is specifically as follows:

[0073]

[0074] Among them, x ij is the observed value of the jth evaluation object on the i-th cost indicator, max{x kj} and min{x kj ) represent the maximum and minimum values ​​of all cost indicators of the jth evaluation object; when x ij When it is close to the minimum value, the membership degree r ij If it is close to 1, it means that the evaluation object performs well on this indicator.

[0075] The calculation method of the benefit-based membership function generated by the application is:

[0076]

[0077] Among them, y ij is the observed value of the jth evaluation object on the i-th benefit indicator, max{y kj} and min{y kj} respectively represent the maximum and minimum values ​​of all cost indicators in the jth evaluation object; when y ij When it approaches the maximum value, the membership degree r ij If it is close to 1, it means that the evaluation object performs well on this indicator.

[0078] After completing the above calculations, determine the evaluation matrix and get U i The fuzzy evaluation matrix is ​​as follows:

[0079]

[0080] Build a fuzzy evaluation model for data value assessment; in the evaluation matrix R of each indicator at the indicator layer where the sub-factors are located i After the fuzzy matrix calculation, the membership vector C′ of the sub-factor index relative to the evaluation set is solved i , C′ i =W i ·R i (i=1,2...6), where W i is the weight of the ith indicator of the three-level indicators, R iis the fuzzy relationship matrix of the i-th indicator of the secondary indicator; similarly, the comprehensive evaluation result of the secondary evaluation layer obtained by the hierarchical analysis method is:

[0081]

[0082]

[0083] Among them, B′1 and B′2 are the comprehensive evaluation results of the secondary evaluation layer, W′ i is the weight of the ith indicator of the secondary indicator, and the comprehensive evaluation result A′ of the primary evaluation layer is:

[0084]

[0085] Where W″ is the weight of the first-level indicator; the comprehensive score of the fuzzy evaluation of data value is calculated, and the score N is scored according to the percentage system according to the evaluation of the comment set:

[0086] N=A′.G T

[0087] Among them, G T It is the membership vector of the comment set level, which is usually used to represent the scores of different levels. According to the scoring criteria, the value of data services can be divided into high, medium and low levels. Grading the data value; according to the pre-determined data value level, the data is graded with reference to the data value.

[0088] Provide typical users with the data service type with the highest comprehensive score; since each industry has different scoring situations for different data service application effects, further, for a certain industry, compare the comprehensive value scores of different data services for the industry, and recommend the data service type with the highest comprehensive score and grade to typical users; the present invention can predict the application effect of data services for users who lack application score data based on historical data using a multi-layer perceptron, and quantify the actual value of the data assets of the supplier enterprise using a hierarchical analysis model and a fuzzy value evaluation method; by analyzing the application effects of different categories of data services on typical users, combined with data service component analysis, the data value is graded, and targeted data services and improvement suggestions are provided to each industry user to help them increase their profits and promote the effective use of data services.

[0089] Example 3, reference Figure 3 , is an embodiment of the present invention, and provides a semantic-based intelligent logistics customer service processing method and system, including a data collection module, a score prediction module, and a comprehensive evaluation module.

[0090] The data collection module is used to collect data on different users' scores for data service application effect indicators and indicator data that characterize user characteristics; the score prediction module is used to predict the scores of application effects of different data services for users with different characteristics based on the multi-layer perceptron model; the comprehensive evaluation module is used to construct a comprehensive evaluation index system to obtain the comprehensive scores of different users for different data services, and provide typical users with the data service type with the highest comprehensive score.

[0091] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0092] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0093] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0094] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A semantic-based intelligent logistics customer service processing method, characterized in that: include: Collect data on different users' scores on data service application effect indicators and indicator data that characterize user characteristics; Based on the multi-layer perceptron model, the scores of application effects of different data services for users with different characteristics are predicted; Build a comprehensive evaluation index system to obtain the comprehensive scores of different users for different data services, and provide typical users with the data service type with the highest comprehensive score; Set up user interaction functions to interact with the platform through natural language instructions, parse natural language instructions and generate corresponding standardized input file instructions and visualization instructions of model results; Set up code generation to generate scripts that adjust model input files, detect logical defects in the files, generate test scenarios, and visualize results.

2. The semantic-based intelligent logistics customer service processing method according to claim 1, characterized in that: The data on the scores of data service application effect indicators by different users includes the data obtained by distributing questionnaires, requiring typical industry users to score the application effects of existing data services, and the application effects include basic data application effects, conventional data application effects, and advanced data application effects.

3. The semantic-based intelligent logistics customer service processing method according to claim 2, characterized in that: The indicator data describing user characteristics includes collecting indicators describing typical users, including basic information, transaction behavior, and benefit and efficiency related indicators.

4. The semantic-based intelligent logistics customer service processing method according to claim 3, characterized in that: The multi-layer perceptron model includes an indicator that describes the user situation as an independent variable and an application effect score as a dependent variable. The multi-layer perceptron MLP is used to predict the evaluation scores of users with different characteristics on the application effect of data services. The architecture of the neural network is designed according to the actual data situation, including the number of input layers, hidden layers, and output layers and the number of nodes in each layer.

5. The semantic-based intelligent logistics customer service processing method according to claim 4, characterized in that: The scoring situation includes applying the trained multi-layer perceptron model to the data set, inputting the values ​​of relevant indicators for typical users, and obtaining the application effect score of a single data service for the user.

6. The semantic-based intelligent logistics customer service processing method according to claim 5, characterized in that: The construction of the comprehensive evaluation index system includes evaluating the value of data assets from two aspects, namely, data asset cost and actual application effect, based on the hierarchical analysis method and the fuzzy evaluation model, constructing a data asset value evaluation index system, and evaluating the actual value of data assets.

7. The semantic-based intelligent logistics customer service processing method according to claim 6, characterized in that: Providing the data service type with the highest comprehensive score to typical users includes obtaining the comprehensive value scores of different users for different data services, comparing the comprehensive value scores of different data services to users, and recommending the data service type with the highest comprehensive value score and level to typical users to help users optimize their decisions.

8. A system using the semantic-based intelligent logistics customer service processing method as claimed in any one of claims 1 to 7, characterized in that: Including data collection module, score prediction module, and comprehensive evaluation module; The data collection module is used to collect data on different users' scores on data service application effect indicators and indicator data describing user characteristics; The score prediction module is used to predict the scores of application effects of different data services for users with different characteristics based on a multi-layer perceptron model; The comprehensive evaluation module is used to construct a comprehensive evaluation index system to obtain comprehensive scores of different users for different data services, and to provide typical users with the data service type with the highest comprehensive score.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the semantic-based intelligent logistics customer service processing method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the semantic-based intelligent logistics customer service processing method described in any one of claims 1 to 7 are implemented.

Images