CN115659011A - Decision case module recommendation implementation method based on autonomous recommendation mechanism - Google Patents
Decision case module recommendation implementation method based on autonomous recommendation mechanism Download PDFInfo
- Publication number
- CN115659011A CN115659011A CN202211259610.4A CN202211259610A CN115659011A CN 115659011 A CN115659011 A CN 115659011A CN 202211259610 A CN202211259610 A CN 202211259610A CN 115659011 A CN115659011 A CN 115659011A
- Authority
- CN
- China
- Prior art keywords
- recommendation
- case
- decision
- event
- scheme
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a decision case module recommendation implementation method based on an autonomous recommendation mechanism, which comprises the following steps: step1, generating a decision case template based on a basic database according to an event processing mechanism, and carrying out classification management on the decision case module to form a case template platform; step2, extracting a specific decision case template from the case template platform according to the event condition to obtain a recommendation scheme to be verified; and Step3, comparing and evaluating the recommended scheme to be verified through an autonomous recommendation mechanism, and obtaining an optimal recommended scheme. The invention solves the problems of low efficiency of manual browsing and selection of the decision case in the urban management process and difficult selection of the decision case due to industry barriers in the urban management process.
Description
Technical Field
The invention relates to the field of public infrastructure decision case templates and event handling, in particular to a decision case module recommendation implementation method based on an autonomous recommendation mechanism.
The invention relates to an invention name.
Background
With the continuous and high-speed development of information technology in China, smart cities are in reality from the concept, and the basic elements of smart cities are various internet of things sensors in the public infrastructure of modern cities, and the sensors are installed at various positions of the cities to monitor various physical signs of the cities in real time and report real-time monitoring data. Hundreds of millions of data information are collected to a monitoring platform to form a big data center, various actual combat application scenes are derived based on the basic data, core scenes are compared, for example, event rule definition is carried out on the data collected by monitoring, countless event information is formed aiming at the monitoring data, then various preposed conditions and verification rules (meeting policy documents and laws and regulations) are added to the events, and finally an event handling case template for reference of a city manager is formed.
After case templates are formed, an urban manager cannot select the case template which best meets all trades from a large number of case templates for use, the efficiency is low firstly when the case templates are manually browsed and selected, and secondly, if the research on a certain trade is not deep enough, the selected case template is not suitable for use; in order to further facilitate the city manager to accurately select and use the case template so as to improve the comprehensive management efficiency of the city, a set of autonomous event case template recommendation mechanism based on the operation status of various industries in the city is developed.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a decision case module recommendation implementation method based on an autonomous recommendation mechanism, so as to solve the problems of low efficiency of manual browsing and selection of decision cases in the urban management process and difficulty in selection of decision cases due to industry barriers in the urban management process.
The invention provides a decision case module recommendation implementation method based on an autonomous recommendation mechanism, which comprises the following steps:
step1, generating a decision case template based on a basic database according to an event processing mechanism, and carrying out classification management on the decision case module to form a case template platform;
step2, extracting a specific decision case template from the case template platform according to the event condition to obtain a recommendation scheme to be verified;
and Step3, comparing and evaluating the recommended scheme to be verified through an autonomous recommendation mechanism, and obtaining an optimal recommended scheme.
Preferably, the autonomous recommendation mechanism comprises a recommendation algorithm, model training, strategy design and evaluation analysis;
the recommendation algorithm learns the recommendation scheme to be verified, obtains the recommendation scheme to be trained, and generates a corresponding recommendation model;
the model training carries out simulation reduction on the recommended scheme to be trained based on the recommendation model to obtain a simulation event case model;
the strategy design is used for screening and sequencing the simulation event case models to obtain a recommendation scheme to be evaluated;
and the evaluation analysis carries out reverse evaluation on the recommended scheme to be evaluated to obtain the optimal recommended scheme.
Preferably, the basic database analyzes and processes the basic data of different industries; the event processing mechanism defines and manages different scene event processing rules in different industries.
Preferably, the extraction of the recommendation to be verified is a function fitting the historical event decision case and the user satisfaction, and the function comprises three dimensional variables:
the content is as follows: the event case content is diversified, and how to extract the characteristics of different event case types needs to be considered for making recommendations;
the user: how to extract the characteristics of the user to which the adopted event case belongs;
scene: event case preferences currently required by users are in various industries, in various scenarios.
Preferably, the model training is performed based on operation data processing of historical decision event cases, the recommendation model is updated according to collected data processing, meanwhile, an online server is used for recording real-time feature import file queues and importing the complete data of the cluster splicing users, and a simulation sample event case is constructed in combination with the recommendation scheme to be trained; and then updating parameters of the recommended model according to the simulation sample event case, and performing on-line model operation to obtain a simulation event case model.
Preferably, the parameters of the recommendation model are stored in a high-performance server cluster and comprise massive original features and vector features.
Preferably, the recommendation algorithm includes Logistic Regression, factorization Machine, GBDT (Gradient Boosting Decision Tree);
the Logistic Regression algorithm realizes automatic classification of the recommended scheme to be verified by constructing a classification model, and obtains classification characteristic data;
the Factorization Machine algorithm combines the classification characteristic data to obtain a decision case to be recommended;
the GBDT (Gradient Boosting Decision Tree) algorithm is used for fitting the Decision case to be recommended based on actual requirements to obtain a predicted value close to a true value, and the Decision case to be recommended with a high predicted value is defined as a recommendation scheme to be trained.
Preferably, the strategy design is based on a recall module to screen and sort the simulation event case models, generate a recommendation list and realize online real-time update.
Preferably, the recall module includes scene deduplication, diversity control and weighting modes.
Preferably, the evaluation analysis evaluates the rationality and the correctness of the recommendation scheme to be evaluated in a reverse deduction mode, outputs an application analysis report to the recommendation scheme to be evaluated according to a set evaluation system, scores the recommendation scheme to be evaluated, and determines an optimal recommendation scheme according to a preset recommendation rank limit.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention can effectively improve the production efficiency of event decision cases in the urban management process.
2. The invention can effectively break the industry barrier of event decision case use in the urban management process.
3. The method can effectively shorten the event decision time in city treatment, improve the event handling efficiency and efficiently enable the smart city.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic diagram of a logic flow structure of a decision case module recommendation implementation method based on an autonomic recommendation mechanism according to the present invention;
FIG. 2 is a schematic diagram of an operation flow structure of an event case of the autonomic recommendation mechanism of the present invention;
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
As shown in fig. 1, the present invention provides a decision case module recommendation implementation method based on an autonomous recommendation mechanism, which includes the following steps:
step1, generating a decision case template based on a basic database according to an event processing mechanism, and carrying out classification management on the decision case module to form a case template platform;
and the basic database performs data analysis processing on basic data of different industries. The basic data of different industries include but are not limited to: mass basic data and related technical parameters of mass equipment in different industries;
the event processing mechanism defines and manages the event processing rules of different scenes in different industries. Wherein, the event processing rule includes but is not limited to: and event handling laws and regulations, behavior specifications and the like defined by various scenes of various industries are processed.
Specifically, mass basic data and related technical parameters of mass equipment in different industries are accessed to a set of Internet of things platform for aggregation, calculation and management to form a basic database required by decision case template generation. Defining and managing the event rules of various industries, such as public security, fire protection, traffic, city transportation and other large industries, and defining a large number of event rules for various subdivided industrial places such as markets, airports, hotels, schools, institutions and the like; defining event processing rules according to early warning rules of scenes such as typhoons, river water levels, smoke concentration and the like; and an event processing mechanism is incorporated for classification specification management. And intelligently producing decision case templates of a large number of events based on the two conditions, and managing the decision case templates to form a case template platform.
Step2, extracting a specific decision case template from a case template platform according to the event condition to obtain a recommendation scheme to be verified;
specifically, the extraction of the recommendation scheme to be verified is a function fitting the historical event decision case and the user satisfaction, and the function comprises three dimensional variables:
the content is as follows: the event case content is diversified, and how to extract the characteristics of different event case types needs to be considered for making recommendations.
The user: how to extract features of users to which adopted event cases belong
Scene: event case preferences currently required by users are in various industries, in various scenarios.
Step3, comparing and evaluating the recommendation scheme to be verified through an autonomous recommendation mechanism to obtain an optimal recommendation scheme;
the self-recommendation mechanism comprises a recommendation algorithm, model training, strategy design and evaluation analysis;
the recommendation algorithm learns the recommendation scheme to be verified, obtains the recommendation scheme to be trained, and generates a corresponding recommendation model;
specifically, the recommended algorithm includes three algorithms of Logistic Regression, factorization Machine, and GBDT (Gradient Boosting Decision Tree);
the Logistic Regression algorithm realizes automatic classification of the recommended scheme to be verified by constructing a classification model, and obtains classification characteristic data; in particular, logistic Regression is a machine learning method for solving the problem of two-classification (0 or 1) to estimate the likelihood of something. The algorithm is used for a classification module of a recommendation system, is mainly used for solving the problem of multi-classification, constructs various classification models required by the system through the algorithm, and supervises the system to automatically learn so as to realize the automation of classification.
Combining the classification characteristic data by a Factorization Machine algorithm to obtain a decision case to be recommended; specifically, the mechanism: the algorithm is to add second-order (or higher-order) feature intersection on the basis of a common linear model, and map a weight matrix of n x n into a space of n x k by using the idea of matrix decomposition. The algorithm is used for combining the atomic decision cases of the system according to the characteristics to generate a new decision case, namely the decision case to be recommended for recommendation.
And fitting the Decision case to be recommended by a GBDT (Gradient Boosting Decision Tree) algorithm based on actual requirements to obtain a predicted value close to a true value, and defining the Decision case to be recommended with a high predicted value as a recommendation scheme to be trained. Specifically, GBDT (Gradient Boosting Decision Tree): the algorithm is an integrated algorithm based on a decision tree. Wherein, the Gradient Boosting is an algorithm in the Boosting of the integration method, and a new learner is iterated through Gradient descent. And the CART decision tree is adopted in the GBDT. The core function of the algorithm in the system is to continuously train a predicted value closest to a true value by using a negative gradient approximation residual concept, namely to acquire a decision case with the highest similarity and recommend the decision case to a user.
The three algorithms are mutually independent in implementation and are mutually related in service, the Logistic Regression algorithm serves the classification module to obtain classification characteristic data, and the Factorization Machine algorithm combines the classification characteristic data according to characteristics to generate a decision case to be recommended. And the GBDT algorithm fits the user requirements to the system to obtain a predicted value close to the true value, and finally selects a recommendation scheme to be trained to recommend to the user.
Model training is carried out on the recommendation scheme to be trained, simulation reduction is carried out on the basis of a recommendation model, and a simulation event case model is obtained;
specifically, model training is performed based on operation data processing of clicking, showing, sharing and the like of historical decision event cases, a recommendation model is updated according to collected data processing, meanwhile, an online server is used for recording real-time feature import file queues, importing complete data of cluster splicing users, and a recommendation scheme to be trained is combined to construct a simulation sample event case. And then updating parameters of the recommended model according to the simulation sample event case, and performing on-line model operation to obtain a simulation event case model. In particular, the parameters of the recommendation model are stored in a high-performance server cluster and comprise massive original features and vector features.
The strategy design is used for screening and sequencing the simulation event case models to obtain a recommendation scheme to be evaluated;
specifically, the strategy design is based on a recall module to screen and sort the simulation event case models, generate a recommendation list and realize online real-time updating. The recall module comprises modes of scene duplicate removal, diversity control, weighting and the like. A very important strategy, the recall strategy, is used in the design of recommendation strategies. The strategy aims to screen a small part of event decision cases meeting the requirements from a large number of event decision cases.
In the implementation steps, the data extraction needs to be implemented based on deep computation, and specifically, data extraction such as event cases, event rules, laws and regulations, policy documents, industries, scenes and the like uses two modes of stream computation and batch computation. And a large-scale and high-performance storage system is used for supporting reading and writing of mass events, and the whole event decision case recommendation system is calculated according to the data.
And performing evaluation analysis to perform reverse evaluation on the recommendation scheme to be evaluated to obtain the optimal recommendation scheme.
Specifically, the evaluation analysis evaluates the reasonability and correctness of the recommendation scheme to be evaluated in a reverse deduction mode, and simultaneously outputs an application analysis report according to a set evaluation system, so as to elaborate the case applicability and recommendation level in detail; meanwhile, scoring is carried out on the recommendation scheme to be evaluated, and the optimal recommendation scheme is determined according to the preset recommendation ranking limit. If the preset recommendation ranking limit is 3, the case with the score of 3 at the top is recommended to the manager for decision.
In practical application, whether the final recommendation effect of the optimal recommendation scheme meets the requirements of users or not needs to be subjected to complex evaluation and analysis, so that the rationality and correctness of the recommendation scheme, the algorithm, the strategy and the calculation are reversely deduced. The factors which can influence the recommendation effect firstly are the change of the candidate content set, the addition and the improvement of the recall module, the addition of the recommendation characteristic, the improvement of the recommendation system architecture, the optimization of the algorithm parameter and the change of the rule strategy. Second, the evaluation requires a complete evaluation system: synthesizing the comprehensive indexes as much as possible into a unique evaluation index, a relatively strong experiment platform and an easy-to-use experiment analysis tool. The final evaluation requires attention to the following points: paying attention to short-term indexes and long-term indexes, paying attention to the influence of synergistic effect and thoroughly carrying out statistical isolation when necessary.
In summary, the precondition of the present invention includes mass sensor monitoring data, a large number of event rules (e.g. emergency handling rules, various equipment monitoring and early warning rules, etc.) in each industry and a large number of event case templates in multiple industries, which are intelligently generated by the above two elements. Secondly, the autonomous recommendation mechanism contains a set of core recommendation algorithm, the recommendation algorithm takes industries as a large class and scenes as a small class, collision calculation is carried out on the recommendation algorithm and various existing event rules in the system, the event templates which are actually adopted by various industries are searched and compared (a behavior analysis algorithm and big data retrieval are adopted) to meet the system recommendation scheme, the recommendation algorithm is operated, training is carried out according to a recommendation model, depth calculation is carried out by using a designed recommendation strategy, and evaluation is carried out by adopting an evaluation system, a plurality of case templates are obtained after the execution of a series of schemes, algorithms, strategies, training and evaluation full flow is finished, an application analysis report aiming at the industries and the scenes is given out aiming at each case template, the case applicability degree and the recommendation level are elaborated in detail; and finally, scoring the case templates, and recommending the case with the highest score to a city manager for decision use. Therefore, a set of relatively perfect self-recommendation mechanism is formed, the self-recommendation mechanism comprises the recommendation scheme, the algorithm, the recommendation model, the strategy, the logic calculation and evaluation system, all the elements work together, and therefore the event case meeting the actual scene needs of the user is searched in the system and finally recommended to the user for reference.
In practical application, as shown in fig. 2, an application scenario is selected by selecting an industry, a demand direction is determined, and a decision case template is generated by a case template platform by adjusting complexity (generally set to 50-100). Then, carrying out comparison and evaluation on the autonomous recommendation mechanism on the decision case template through the decision case module recommendation implementation method based on the autonomous recommendation mechanism, and obtaining an optimal recommendation scheme according to the passing of the review analysis report so as to implement and use; conversely, decision case templates that fail in reviewing the analysis report are discarded. Further, after the optimal recommendation scheme is put into use, whether the recommendation scheme needs to be reviewed and evaluated in a batch again according to the autonomous recommendation mechanism can be evaluated according to the actual use condition. Therefore, in bidirectional autonomous review, evaluation and learning based on historical data and actual use conditions, repeated learning of an autonomous recommendation mechanism is promoted by continuously updating data on line, and the final evaluation value of the decision case module recommendation implementation method based on the autonomous recommendation mechanism has higher reference application value.
Further, in practical application, when a certain responsible department is unknown to the governed business process or the event handling process is ambiguous, the event case autonomous recommendation mechanism can be used to refer to mature event case handling processes of other cities or departments, and excellent and mature similar event case handling experiences can be absorbed to enable the department.
On the other hand, when the responsibility department is not very familiar with the industry specifications, laws and regulations, policy documents and the like and has a knowledge blind area so as to influence the selection of the event cases and the decision of major events, a plurality of specific event cases with higher scores can be selected through an event case self-recommending mechanism, and the analysis report after the self-recommended event cases are subjected to statistical induction can be subjected to point-to-point proofreading with the industry specifications, laws and regulations and policy documents. Therefore, the time for compiling, examining and testing the event cases is saved, and the city management operation efficiency is improved.
It is well within the knowledge of a person skilled in the art to implement the system and its various devices, modules, units provided by the present invention in a purely computer readable program code means that the same functionality can be implemented by logically programming method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units included in the system for realizing various functions can also be regarded as structures in the hardware component; means, modules, units for performing the various functions may also be regarded as structures within both software modules and hardware components for performing the method.
The foregoing description has described specific embodiments of the present invention. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A decision case module recommendation implementation method based on an autonomous recommendation mechanism is characterized by comprising the following steps:
step1, generating a decision case template based on a basic database according to an event processing mechanism, and carrying out classification management on the decision case module to form a case template platform;
step2, extracting a specific decision case template from the case template platform according to the event condition to obtain a recommendation scheme to be verified;
and Step3, comparing and evaluating the recommended scheme to be verified through an autonomous recommendation mechanism, and obtaining an optimal recommended scheme.
2. The method for implementing decision case module recommendation based on autonomous recommendation mechanism of claim 1, wherein the autonomous recommendation mechanism comprises recommendation algorithm, model training, strategy design, evaluation analysis;
the recommendation algorithm learns the recommendation scheme to be verified, obtains the recommendation scheme to be trained, and generates a corresponding recommendation model;
the model training carries out simulation reduction on the recommended scheme to be trained based on the recommendation model to obtain a simulation event case model;
the strategy design is used for screening and sequencing the simulation event case models to obtain a recommendation scheme to be evaluated;
and the evaluation analysis carries out reverse evaluation on the recommended scheme to be evaluated to obtain the optimal recommended scheme.
3. The decision case module recommendation implementation method based on the autonomous recommendation mechanism as claimed in claim 1, wherein the basic database performs data analysis processing on basic data of different industries; the event processing mechanism defines and manages different scene event processing rules in different industries.
4. The method for implementing case-based module recommendation based on autonomic recommendation mechanism as claimed in claim 1, wherein the extraction of the recommendation scheme to be verified is a function fitting the historical event decision cases and the user satisfaction, the function comprising three dimensional variables:
the content is as follows: the event case content is diversified, and the method needs to consider how to extract the characteristics of different event case types and recommend the characteristics;
the user: how to extract the characteristics of the user to which the adopted event case belongs;
scene: event case preferences currently required by users are in various industries, in various scenarios.
5. The method for implementing module recommendation of decision cases based on an autonomous recommendation mechanism as claimed in claim 2, wherein the model training is based on the operation data processing of historical decision event cases, the recommendation model is updated according to the collected data processing, meanwhile, an online server is used to record the complete data of the real-time feature import file queue import cluster splicing users, and a simulation sample event case is constructed in combination with the recommendation scheme to be trained; and then updating parameters of the recommendation model according to the simulation sample event case, and performing on-line model operation to obtain a simulation event case model.
6. The method as claimed in claim 5, wherein the parameters of the recommendation model are stored in a high performance server cluster, and include a large number of original features and vector features.
7. The decision case module recommendation implementation method based on the autonomous recommendation mechanism of claim 2, wherein the recommendation algorithm comprises Logistic Regression, factitious Machine, GBDT;
the Logistic Regression algorithm realizes automatic classification of the recommended scheme to be verified by constructing a classification model, and obtains classification characteristic data;
the Factorization Machine algorithm combines the classification characteristic data to obtain a decision case to be recommended;
and the GBDT algorithm is used for fitting the decision-making case to be recommended based on actual requirements to obtain a predicted value close to a true value, and the decision-making case to be recommended with a high predicted value is defined as a recommendation scheme to be trained.
8. The method as claimed in claim 2, wherein the strategy design is based on a recall module to filter and sort the simulation event case models, generate a recommendation list, and implement online real-time update.
9. The method for implementing module recommendation for decision cases based on autonomic recommendation mechanism as claimed in claim 8, wherein said recall module comprises scene de-duplication, diversity control, weighting.
10. The decision case module recommendation implementation method based on the autonomic recommendation mechanism as claimed in claim 2, wherein the evaluation analysis evaluates the reasonableness and correctness of the recommendation scheme to be evaluated in a reverse deduction manner, outputs an application analysis report on the recommendation scheme to be evaluated according to a given evaluation system, scores the recommendation scheme to be evaluated, and determines an optimal recommendation scheme according to a preset recommendation rank limit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211259610.4A CN115659011A (en) | 2022-10-14 | 2022-10-14 | Decision case module recommendation implementation method based on autonomous recommendation mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211259610.4A CN115659011A (en) | 2022-10-14 | 2022-10-14 | Decision case module recommendation implementation method based on autonomous recommendation mechanism |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115659011A true CN115659011A (en) | 2023-01-31 |
Family
ID=84987216
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211259610.4A Pending CN115659011A (en) | 2022-10-14 | 2022-10-14 | Decision case module recommendation implementation method based on autonomous recommendation mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115659011A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829061A (en) * | 2023-02-21 | 2023-03-21 | 中国电子科技集团公司第二十八研究所 | Emergency accident disposal method based on historical case and empirical knowledge learning |
| CN116993296A (en) * | 2023-08-15 | 2023-11-03 | 深圳市中联信信息技术有限公司 | Intelligent supervision management system and method applied to engineering design interaction platform |
| CN117035125A (en) * | 2023-10-09 | 2023-11-10 | 中电科大数据研究院有限公司 | Decision platform, method and device for distributed artificial intelligence model |
| CN117196399A (en) * | 2023-09-21 | 2023-12-08 | 深圳市科荣软件股份有限公司 | Customer service center operation supervision optimization system based on data analysis |
| CN117350288A (en) * | 2023-12-01 | 2024-01-05 | 浙商银行股份有限公司 | Case matching-based network security operation auxiliary decision-making method, system and device |
| CN117196399B (en) * | 2023-09-21 | 2024-05-31 | 深圳市科荣软件股份有限公司 | Customer service center operation supervision optimization system based on data analysis |
-
2022
- 2022-10-14 CN CN202211259610.4A patent/CN115659011A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829061A (en) * | 2023-02-21 | 2023-03-21 | 中国电子科技集团公司第二十八研究所 | Emergency accident disposal method based on historical case and empirical knowledge learning |
| CN116993296A (en) * | 2023-08-15 | 2023-11-03 | 深圳市中联信信息技术有限公司 | Intelligent supervision management system and method applied to engineering design interaction platform |
| CN116993296B (en) * | 2023-08-15 | 2024-04-16 | 深圳市中联信信息技术有限公司 | Intelligent supervision management system and method applied to engineering design interaction platform |
| CN117196399A (en) * | 2023-09-21 | 2023-12-08 | 深圳市科荣软件股份有限公司 | Customer service center operation supervision optimization system based on data analysis |
| CN117196399B (en) * | 2023-09-21 | 2024-05-31 | 深圳市科荣软件股份有限公司 | Customer service center operation supervision optimization system based on data analysis |
| CN117035125A (en) * | 2023-10-09 | 2023-11-10 | 中电科大数据研究院有限公司 | Decision platform, method and device for distributed artificial intelligence model |
| CN117035125B (en) * | 2023-10-09 | 2024-01-02 | 中电科大数据研究院有限公司 | Decision system, method and device of distributed artificial intelligence model |
| CN117350288A (en) * | 2023-12-01 | 2024-01-05 | 浙商银行股份有限公司 | Case matching-based network security operation auxiliary decision-making method, system and device |
| CN117350288B (en) * | 2023-12-01 | 2024-05-03 | 浙商银行股份有限公司 | Case matching-based network security operation auxiliary decision-making method, system and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Debrah et al. | Artificial intelligence in green building | |
| CN115659011A (en) | Decision case module recommendation implementation method based on autonomous recommendation mechanism | |
| Thangavel et al. | Student placement analyzer: A recommendation system using machine learning | |
| CN107729915A (en) | For the method and system for the key character for determining machine learning sample | |
| CN112800229B (en) | Knowledge graph embedding-based semi-supervised aspect-level emotion analysis method for case-involved field | |
| Seebacher | Predictive intelligence for data-driven managers | |
| Kour et al. | Analysis of student performance using Machine learning Algorithms | |
| Arafath et al. | Predicting career using data mining | |
| Shi et al. | [Retracted] A BP Neural Network‐Based Early Warning Model for Student Performance in the Context of Big Data | |
| Matetic | Mining learning management system data using interpretable neural networks | |
| Kumar et al. | Relevance of data mining techniques in edification sector | |
| Feng | Data analysis and prediction modeling based on deep learning in E-commerce | |
| Caraka | Prediction of Euro 50 Using Back Propagation Neural Network (BPNN) and Genetic Algorithm (GA) | |
| Mostafa et al. | Application of Artificial Intelligence Tools with BIM Technology in Construction Management: Literature | |
| TWI833098B (en) | Intellectual quality management method, electronic device and readable storage medium | |
| Kumar et al. | Selection of evolutionary approach based hybrid data mining algorithms for decision support systems and business intelligence | |
| Krömer et al. | Mining traffic accident features by evolutionary fuzzy rules | |
| Tan et al. | Predicting dropout from online education based on neural networks | |
| Arabikhan | Telecommuting choice modelling using fuzzy rule based networks | |
| CN117390522B (en) | Online deep learning level prediction method and device based on process and result fusion | |
| Nuankaew et al. | Predictive Model for Clustering Learning Outcomes Affected by COVID-19 Using Ensemble Learning Techniques. | |
| Kubaščíková et al. | Using of non-financial data in predictive models | |
| Wu | Multi-task achievement prediction model based on grade change trend | |
| Zhou | Construction of genetic algorithm model for fitness program optimization of middle school students | |
| Chakrabarty et al. | Tools and Methods of Educational Data Mining: A Review |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |