CN112153636A - Method for predicting number portability and roll-out of telecommunication industry user based on machine learning - Google Patents
Method for predicting number portability and roll-out of telecommunication industry user based on machine learning Download PDFInfo
- Publication number
- CN112153636A CN112153636A CN202011178646.0A CN202011178646A CN112153636A CN 112153636 A CN112153636 A CN 112153636A CN 202011178646 A CN202011178646 A CN 202011178646A CN 112153636 A CN112153636 A CN 112153636A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- prediction
- predicting
- prediction model
- samples
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000010801 machine learning Methods 0.000 title claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 20
- 238000012360 testing method Methods 0.000 claims abstract description 9
- 238000005457 optimization Methods 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000011156 evaluation Methods 0.000 claims abstract description 7
- 238000005070 sampling Methods 0.000 claims abstract description 4
- 238000004140 cleaning Methods 0.000 claims description 5
- 230000006399 behavior Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 2
- 239000013589 supplement Substances 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000012423 maintenance Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 6
- 238000012546 transfer Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000586 desensitisation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000013077 scoring method Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000007473 univariate analysis Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W8/00—Network data management
- H04W8/26—Network addressing or numbering for mobility support
- H04W8/28—Number portability ; Network address portability
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention relates to the technical field of telecommunication, in particular to a method for predicting number portability and transfer-out of telecommunication users based on machine learning, which comprises the following steps: 1) collecting characteristic variable data, preprocessing the characteristic variable data, storing the characteristic variable data in a database, sampling samples in the database, and controlling the proportion of positive samples to negative samples to be 1: 10; 2) samples are randomly divided into a training set and a testing set; 3) selecting an XGboost algorithm as a basis to construct a prediction model, inputting a training set to train the prediction model, and obtaining a prediction probability value and the importance degree of characteristics; 4) and performing data prediction on the test set by using the trained model, evaluating the prediction model according to the prediction result, and performing optimization iteration on the prediction model if the evaluation result is lower than a threshold value. The invention has the beneficial effects that: the prediction efficiency is improved, early warning and timely maintenance are achieved in advance, and the prediction model can dynamically perform optimization iteration.
Description
Technical Field
The invention relates to the technical field of telecommunication, in particular to a method for predicting number portability and transfer-out of telecommunication users based on machine learning.
Background
The number portability service is implemented completely in 11 months in 2019 according to the requirements of the Ministry of industry and communications, the main content is that a user can select a proper telecom operator according to own will, and meanwhile, in the process, the original number can be reserved, and the purpose of moving a mobile phone without changing the number is achieved.
The number portability to the operator is actually divided into two parts: and transferring out the number portability and transferring in the number portability. The number portability is that the user carries the original number to be transferred from the local network operator to other network operators, and can be regarded as a situation that the high-risk user is off the network. The sign-on transfer is opposite.
The customer resources are the core competitiveness of telecom operation enterprises, and how to reduce customer loss, reduce the probability of number portability and transfer out of customers and reduce the economic loss caused by number portability and transfer out of customers becomes a main topic discussed by the telecom operation enterprises. The telecommunication enterprises actively utilize the leading-edge technology and capital, so that the enterprises develop towards intellectualization, synthesis and individuation, and the competitiveness capability is improved, so as to maximize the market share and profit. The method aims to solve the problems of reduced market share and reduced income caused by the user number portability transferring, and simultaneously aims to improve the success rate of saving, reduce the number portability transferring rate and reduce the income loss caused by the user number portability transferring.
Before the number portability network is developed, due to the lack of forward samples, namely users who actually carry numbers to roll out, most of the established models are based on the traditional data mining method. Most of the models are non-machine learning models, such as rule empirical model analysis methods and expert scoring methods. Through several network switching scenes made by the service side, the statistics processing, the analysis and the induction are carried out, and then different network switching probabilities are divided. The method has the problems that the method is easy to approximate and has a plurality of uncertain factors, so that the prediction accuracy of the network forwarding user is not high, and the advance warning of the number portability forwarding of the user is difficult to realize.
Disclosure of Invention
The invention aims to overcome the defects and provide a method for predicting number portability and transfer-out of telecommunication users based on machine learning, so that accurate and effective early warning is performed in advance.
The present invention achieves the above objectives by the following desensitization protocol: a method for predicting number portability and roll-out of telecommunication industry users based on machine learning comprises the following steps:
1) collecting characteristic variable data, preprocessing the characteristic variable data, storing the characteristic variable data in a database, sampling samples in the database, and controlling the proportion of positive samples to negative samples to be 1: 10;
2) samples are randomly divided into a training set and a testing set;
3) selecting an XGboost algorithm as a basis to construct a prediction model, inputting a training set to train the prediction model, and obtaining a prediction probability value and the importance degree of characteristics;
4) and performing data prediction on the test set by using the trained model, evaluating the prediction model according to the prediction result, and performing optimization iteration on the prediction model if the evaluation result is lower than a threshold value.
Preferably, the characteristic variables take into account various dimensional characteristics of existing inventory users, including basic attributes, behavior data, package information, consumption characteristics, terminal information and derivative variables.
Preferably, the preprocessing comprises data cleaning and data conversion, wherein the data cleaning comprises correcting error values, filling missing values and normalizing data types.
Preferably, the missing values are supplemented with a median or zero value.
Preferably, a positive sample represents a carry number roll-out user and a negative sample represents a non-carry number roll-out user.
Preferably, five-fold cross validation and network search are adopted to obtain the optimal solution of the algorithm model.
Preferably, the evaluation of the algorithmic model is performed using the F1 values and AUC values.
Preferably, the threshold is set to 0.5.
Preferably, the optimization iteration comprises the following method: the method comprises the steps of dividing negative samples in a training set into N equal parts, combining a positive sample data set and the negative sample data set of each equal part to form N small training sets, training each small training set by using an XGboost algorithm to obtain N base models, and calculating the average value of output values of the N base models.
The invention has the beneficial effects that: compared with the traditional non-machine learning method, the number portability and roll-out condition of the user can be more accurately predicted, and early warning and timely maintenance are achieved in advance; the prediction model can dynamically perform optimization iteration, and the prediction efficiency is improved.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a schematic diagram of the composition of characteristic variables of the present invention;
FIG. 3 is a schematic flow chart of optimization iteration in the method of the present invention.
Detailed Description
The invention is further described below with reference to specific embodiments, but the scope of protection of the invention is not limited thereto:
example (b): as shown in fig. 1, a method for predicting the carrier roll-out of a telecom business user based on machine learning includes the following steps:
1) collecting characteristic variable data, preprocessing the characteristic variable data, storing the characteristic variable data in a database, sampling samples in the database, and controlling the proportion of positive samples to negative samples to be 1: and 10, positive samples represent carry-out users, and negative samples represent non-carry-out users.
Defining the caliber carried out by the target variable, namely a positive sample:
the actual number portability of the user is transferred to the caliber:
taking the target user in the nth month as an example, one of the following requirements is satisfied:
and (n +1 month) or (n +2 month) or (n +3 month) carry number roll-out users.
As shown in fig. 2, the feature variables include features of various dimensions acquired by existing stock users, and are gradually subdivided into 151 subdivided feature variables starting from six dimensions of basic attributes, behavior data, package information, consumption features, terminal information and derived variables of the users, univariate analysis is performed on part of key variables, the relationship between the key variables and the number portability of the users is measured, whether the key variables conform to actual business rules or not is checked, and the required feature variables are acquired and determined.
And carrying out data preprocessing after acquiring the required characteristic variables, wherein the preprocessing comprises data cleaning and data conversion. The missing numerical values are filled by adopting zero values and median numbers, for example, the age of the user in the invention can be filled by adopting the median numbers, and if the charge amount of the user is missing, the zero values are filled. If there are a large number of missing instances of a feature, the ratio exceeding ninety percent, then the feature is removed. And converting category characteristics, such as the terminal model of a mobile phone, the package name of a user and the traffic conversation trend of the user, by using the one-hot code to convert the categories so as to enable the data to be suitable for a matching algorithm model.
2) Samples are randomly divided into a training set and a testing set;
3) selecting an XGboost algorithm as a prediction model, obtaining a loss function and a prediction probability value, and finishing the training of the prediction model, wherein the loss function represents the inconsistency degree between a predicted user to be transferred and an actual user to be transferred;
Wherein,probability value, y, representing the predicted value of the model, i.e. the user portable roll-out predictiontClass label representing nth sample, K representing number of trees, fkRepresenting the K-th tree model,is a loss function that is the degree of disparity between the predicted and actual roll-out users in the present invention,to measure its fit. If the loss function value is smaller, the model robustness is higher, and the fitting effect is better.The sum of the complexity of K trees is the regularization term, namely the complexity of the estimated carry-out model.
t represents the number of leaf nodes per tree, omega represents the set of fractional components of the leaf nodes per tree, and gamma and lambda are adjustable coefficients.
And finally, simplifying and approximating the objective function by using Taylor second-order expansion to obtain an optimal solution.
4) And predicting data of the test set by using the trained prediction model, and evaluating the prediction model according to the prediction result, wherein in the embodiment, the evaluation indexes are F1 values, AUC, recall rate and accuracy rate. If the evaluation index is lower than the threshold, considering the influence of the number of the collected samples and the model features on the model performance, as shown in fig. 3, the model is optimized: the method comprises the steps of dividing negative samples in a training set into N equal parts, combining a positive sample data set and the negative sample data set of each equal part to form N small training sets, training each small training set by using an XGboost algorithm to obtain N base models, and calculating the average value of output values of the N base models.
The various indices of the model on the test set are as follows:
AUC value (Train): 0.966670
Recall (Train): 0.590293
Accuracy (Train): 0.942627
F1 value (Train): 0.724977
A reasonable threshold value is established according to the situation scale given by the service side. If the service side needs to maintain a large number of users, a smaller threshold value can be selected. Or the traffic side may prefer accuracy, a larger threshold may be selected.
While the invention has been described in connection with specific embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (9)
1. A method for predicting number portability and roll-out of telecommunication industry users based on machine learning is characterized by comprising the following steps:
1) collecting characteristic variable data, preprocessing the characteristic variable data, storing the characteristic variable data in a database, sampling samples in the database, and controlling the proportion of positive samples to negative samples to be 1: 10;
2) samples are randomly divided into a training set and a testing set;
3) selecting an XGboost algorithm as a basis to construct a prediction model, inputting a training set to train the prediction model, and obtaining a prediction probability value and the importance degree of characteristics;
4) and performing data prediction on the test set by using the trained model, evaluating the prediction model according to the prediction result, and performing optimization iteration on the prediction model if the evaluation result is lower than a threshold value.
2. The method for predicting the carrier roll-out of the telecommunication industry user based on the machine learning of claim 1, wherein the feature variables comprise various dimensional features of existing stock users, including basic attributes, behavior data, package information, consumption features, terminal information and derivative variables.
3. The method for predicting the carrier roll-out of the telecommunication industry user based on the machine learning as claimed in claim 2, wherein the preprocessing comprises data cleaning and data conversion, the data cleaning comprises correcting error values, and zero value is adopted to supplement missing values.
4. The method for predicting the carrier roll-out of the telecom industry users based on the machine learning as claimed in claim 3, wherein the missing value is supplemented with a median.
5. The method for predicting the carrier roll-out of the telecommunication industry user based on the machine learning as claimed in claim 3, wherein a positive sample represents the carrier roll-out user and a negative sample represents the non-carrier roll-out user.
6. The method for predicting the carrier roll-out of the telecommunication industry user based on the machine learning as claimed in claim 5, wherein the optimal solution of the prediction model is obtained by adopting five-fold cross validation and network search.
7. The method for predicting the carrier roll-out of the telecommunication industry user based on the machine learning of claim 6, wherein the F1 value and the AUC value are used for the evaluation of the prediction model.
8. The method for predicting the carrier roll-out of the telecom industry users based on the machine learning of claim 7, wherein the threshold is set to 0.5.
9. The method for predicting the carrier roll-out of the telecom industry users based on the machine learning of claim 8, wherein the optimization iteration comprises the following steps: the method comprises the steps of dividing negative samples in a training set into N equal parts, combining a positive sample data set and the negative sample data set of each equal part to form N small training sets, training each small training set by using an XGboost algorithm to obtain N base models, and calculating the average value of output values of the N base models.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011178646.0A CN112153636A (en) | 2020-10-29 | 2020-10-29 | Method for predicting number portability and roll-out of telecommunication industry user based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011178646.0A CN112153636A (en) | 2020-10-29 | 2020-10-29 | Method for predicting number portability and roll-out of telecommunication industry user based on machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112153636A true CN112153636A (en) | 2020-12-29 |
Family
ID=73953560
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011178646.0A Pending CN112153636A (en) | 2020-10-29 | 2020-10-29 | Method for predicting number portability and roll-out of telecommunication industry user based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112153636A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115134805A (en) * | 2021-03-29 | 2022-09-30 | 中国移动通信集团福建有限公司 | Method, device, equipment and storage medium for predicting potential carried-in different network numbers |
CN115412421A (en) * | 2022-08-30 | 2022-11-29 | 南京华苏科技有限公司 | Unsatisfactory user early warning method based on CNN-LSTM model |
CN115988475A (en) * | 2022-12-20 | 2023-04-18 | 中国联合网络通信集团有限公司 | Prediction method, equipment and storage medium of portable user |
CN116033370A (en) * | 2021-10-25 | 2023-04-28 | 中国移动通信集团广东有限公司 | Method and device for processing number-carrying network transfer |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107832581A (en) * | 2017-12-15 | 2018-03-23 | 百度在线网络技术(北京)有限公司 | Trend prediction method and device |
CN109344201A (en) * | 2018-10-17 | 2019-02-15 | 国网江苏省电力有限公司信息通信分公司 | A kind of database performance load evaluation system and method based on machine learning |
CN109451527A (en) * | 2018-12-21 | 2019-03-08 | 广东宜通世纪科技股份有限公司 | A kind of mobile communication subscriber is lost day granularity prediction technique and device |
CN109558962A (en) * | 2017-09-26 | 2019-04-02 | 中国移动通信集团山西有限公司 | Predict device, method and storage medium that telecommunication user is lost |
CN109636446A (en) * | 2018-11-16 | 2019-04-16 | 北京奇虎科技有限公司 | Customer churn prediction technique, device and electronic equipment |
CN109886755A (en) * | 2019-03-04 | 2019-06-14 | 深圳微品致远信息科技有限公司 | A kind of communication user attrition prediction method and system based on evolution algorithm |
US20190318202A1 (en) * | 2016-10-31 | 2019-10-17 | Tencent Technology (Shenzhen) Company Limited | Machine learning model training method and apparatus, server, and storage medium |
CN110472817A (en) * | 2019-07-03 | 2019-11-19 | 西北大学 | A kind of XGBoost of combination deep neural network integrates credit evaluation system and its method |
CN110866767A (en) * | 2018-08-27 | 2020-03-06 | 中国移动通信集团江西有限公司 | Method, device, equipment and medium for predicting satisfaction degree of telecommunication user |
US20200120003A1 (en) * | 2018-10-10 | 2020-04-16 | Sandvine Corporation | System and method for predicting and reducing subscriber churn |
CN111092762A (en) * | 2019-12-19 | 2020-05-01 | 深圳市博瑞得科技有限公司 | Prediction method, device and storage medium for number portability potential user |
CN111242358A (en) * | 2020-01-07 | 2020-06-05 | 杭州策知通科技有限公司 | Enterprise information loss prediction method with double-layer structure |
CN111275245A (en) * | 2020-01-13 | 2020-06-12 | 宜通世纪物联网研究院(广州)有限公司 | Potential network switching user identification method, system, message pushing method, device and medium |
CN111582577A (en) * | 2020-05-07 | 2020-08-25 | 北京思特奇信息技术股份有限公司 | Method, system, medium and equipment for predicting off-network of telecommunication user |
-
2020
- 2020-10-29 CN CN202011178646.0A patent/CN112153636A/en active Pending
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190318202A1 (en) * | 2016-10-31 | 2019-10-17 | Tencent Technology (Shenzhen) Company Limited | Machine learning model training method and apparatus, server, and storage medium |
CN109558962A (en) * | 2017-09-26 | 2019-04-02 | 中国移动通信集团山西有限公司 | Predict device, method and storage medium that telecommunication user is lost |
CN107832581A (en) * | 2017-12-15 | 2018-03-23 | 百度在线网络技术(北京)有限公司 | Trend prediction method and device |
CN110866767A (en) * | 2018-08-27 | 2020-03-06 | 中国移动通信集团江西有限公司 | Method, device, equipment and medium for predicting satisfaction degree of telecommunication user |
US20200120003A1 (en) * | 2018-10-10 | 2020-04-16 | Sandvine Corporation | System and method for predicting and reducing subscriber churn |
CN109344201A (en) * | 2018-10-17 | 2019-02-15 | 国网江苏省电力有限公司信息通信分公司 | A kind of database performance load evaluation system and method based on machine learning |
CN109636446A (en) * | 2018-11-16 | 2019-04-16 | 北京奇虎科技有限公司 | Customer churn prediction technique, device and electronic equipment |
CN109451527A (en) * | 2018-12-21 | 2019-03-08 | 广东宜通世纪科技股份有限公司 | A kind of mobile communication subscriber is lost day granularity prediction technique and device |
CN109886755A (en) * | 2019-03-04 | 2019-06-14 | 深圳微品致远信息科技有限公司 | A kind of communication user attrition prediction method and system based on evolution algorithm |
CN110472817A (en) * | 2019-07-03 | 2019-11-19 | 西北大学 | A kind of XGBoost of combination deep neural network integrates credit evaluation system and its method |
CN111092762A (en) * | 2019-12-19 | 2020-05-01 | 深圳市博瑞得科技有限公司 | Prediction method, device and storage medium for number portability potential user |
CN111242358A (en) * | 2020-01-07 | 2020-06-05 | 杭州策知通科技有限公司 | Enterprise information loss prediction method with double-layer structure |
CN111275245A (en) * | 2020-01-13 | 2020-06-12 | 宜通世纪物联网研究院(广州)有限公司 | Potential network switching user identification method, system, message pushing method, device and medium |
CN111582577A (en) * | 2020-05-07 | 2020-08-25 | 北京思特奇信息技术股份有限公司 | Method, system, medium and equipment for predicting off-network of telecommunication user |
Non-Patent Citations (6)
Title |
---|
任新月;: "机器学习在电信客户离网预测中的应用", 信息通信, no. 05 * |
李为康;杨小兵;: "一种基于双层融合结构的客户流失预测模型", 小型微型计算机系统, no. 08 * |
沈江明;张磊;曾志勇;: "基于深度置信神经网络的电信客户流失分析", 通讯世界, no. 06 * |
赵慧;刘颖慧;崔羽飞;张第;: "机器学习在运营商用户流失预警中的运用", 信息通信技术, no. 01 * |
黄展正: "DG电信公司宽带用户流失的预警模型构建", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》, no. 05 * |
龙克树;邓娟;刘晓斌;: "基于机器学习算法的运营商用户流失预判及应对策略研究", 信息记录材料, no. 05 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115134805A (en) * | 2021-03-29 | 2022-09-30 | 中国移动通信集团福建有限公司 | Method, device, equipment and storage medium for predicting potential carried-in different network numbers |
CN116033370A (en) * | 2021-10-25 | 2023-04-28 | 中国移动通信集团广东有限公司 | Method and device for processing number-carrying network transfer |
CN115412421A (en) * | 2022-08-30 | 2022-11-29 | 南京华苏科技有限公司 | Unsatisfactory user early warning method based on CNN-LSTM model |
CN115988475A (en) * | 2022-12-20 | 2023-04-18 | 中国联合网络通信集团有限公司 | Prediction method, equipment and storage medium of portable user |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112153636A (en) | Method for predicting number portability and roll-out of telecommunication industry user based on machine learning | |
CN109492026B (en) | Telecommunication fraud classification detection method based on improved active learning technology | |
CN107766929A (en) | model analysis method and device | |
CN112054943B (en) | Traffic prediction method for mobile network base station | |
CN110309967A (en) | Prediction technique, system, equipment and the storage medium of customer service session grading system | |
CN109787821B (en) | Intelligent prediction method for large-scale mobile client traffic consumption | |
CN112200375B (en) | Prediction model generation method, prediction model generation device, and computer-readable medium | |
CN112149352B (en) | Prediction method for marketing activity clicking by combining GBDT automatic characteristic engineering | |
CN107704868A (en) | Tenant group clustering method based on Mobile solution usage behavior | |
CN114528395A (en) | Risk prediction method for text word feature double-line attention fusion | |
CN112883062A (en) | Self-defined rule checking method not based on rule | |
CN113780345A (en) | Small sample classification method and system facing small and medium-sized enterprises and based on tensor attention | |
CN116245399A (en) | Model training method and device, nonvolatile storage medium and electronic equipment | |
CN116579640A (en) | Power marketing service channel user experience assessment method and system | |
CN105873119A (en) | Method for classifying flow use behaviors of mobile network user groups | |
CN113486174A (en) | Model training, reading understanding method and device, electronic equipment and storage medium | |
CN109543571B (en) | Intelligent identification and retrieval method for special-shaped processing characteristics of complex products | |
CN108763289B (en) | Massive heterogeneous sensor format data analysis method | |
CN110955835A (en) | Sharing platform information publishing system based on big data technology | |
CN114519343A (en) | 95598-based repeated incoming call preprocessing method, device, equipment and storage medium | |
CN109919811B (en) | Insurance agent culture scheme generation method based on big data and related equipment | |
CN114066506A (en) | AI analysis algorithm for network behavior | |
CN113761897A (en) | Text big data-based call center customer service work order entity identification method | |
CN112749841A (en) | User public praise prediction method and system based on self-training learning | |
CN114492552A (en) | Method, device and equipment for training broadband user authenticity judgment model |
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 |