Developing a prediction model for customer churn from electronic banking services using data mining

Background:Given the importance of customers as the most valuable assets of organizations,customer retention seems to be an essential,basic requirement for any organization.Banks are no exception to this rule.The competitive atmosphere within which electronic banking services are provided by different banks increases the necessity of customer retention.Methods:Being based on existing information technologies which allow one to collect data from organizations’databases,data mining introduces a powerful tool for the extraction of knowledge from huge amounts of data.In this research,the decision tree technique was applied to build a model incorporating this knowledge.Results:The results represent the characteristics of churned customers.Conclusions:Bank managers can identify churners in future using the results of decision tree.They should be provide some strategies for customers whose features are getting more likely to churner’s features.

The Research of Customer Loyalty Improvement in Telecom Industry Based on NPS Data Mining

In recent years, the telecommunications have used the concept of NPS(Net Promoter Score) for customer relationship management, but there is neither definite theory research nor instructive instance research. However, this paper summarizes an approach with instance case analysis to improve customer loyalty via NPS data mining, which has extensive and practical significance for tele-companies. First, this paper finds some driven forces of customer loyalty, which are relative to customer consumption such as the call duration, the usage of data, ARPU, etc., by using some innovative reasoning-analysis based on IG(Information Gain) and xg-boost decision-making tree model, so the tele-companies can predict the role of individual customer and form daily monitoring on big data, which will save a lot of NPS survey cost. Second, this paper summarizes how customer group feature impacts the relationship between NPS and financial performance. Taking ARPU value as the performance goals, we divide the sample customers into 6 groups and summarize their characteristics based on k-means clustering, and give targeted suggestion of each group.

Hybrid Data Mining Models for Predicting Customer Churn

The term “customer churn” is used in the industry of information and communication technology (ICT) to indicate those customers who are about to leave for a new competitor, or end their subscription. Predicting this behavior is very important for real life market and competition, and it is essential to manage it. In this paper, three hybrid models are investigated to develop an accurate and efficient churn prediction model. The three models are based on two phases;the clustering phase and the prediction phase. In the first phase, customer data is filtered. The second phase predicts the customer behavior. The first model investigates the k-means algorithm for data filtering, and Multilayer Perceptron Artificial Neural Networks (MLP-ANN) for prediction. The second model uses hierarchical clustering with MLP-ANN. The third one uses self organizing maps (SOM) with MLP-ANN. The three models are developed based on real data then the accuracy and churn rate values are calculated and compared. The comparison with the other models shows that the three hybrid models outperformed single common models.

Aviation Safety and Data Mining in Marketing Dimension

The aviation industry is a sector that is developing, changing and growing every day in terms of technological and legal framework. There are generally three factors that enable airlines to hold on to the market. These factors are safety, service quality and price. Airline companies can analyze the customers in the market with a focus on price and quality and develop a business model according to their expectations. For example, business class and economy class passenger expectations are different from each other, so the service and price to be offered to them will be different. However, all customers have one common expectation and that is safety. No matter how high quality the service is or how cheap the price is, no one wants to fly with an airline or plane that is not safe. From an airline company’s point of view, an accident or breakdown of one of the company’s aircraft can cause irreparable image loss and financial damage. If we look at past examples, we see that there are many airline companies or maintenance organizations that could not recover after an accident and went bankrupt. Safety is an indispensable factor. Therefore, there is a unit in the sector called the safety management system (SMS), which collects data by taking a proactive and reactive approach. The way and purpose of the safety management system is to take a proactive approach to recognize and prevent unsafe situations before they cause accidents or breakdowns, or to take a reactive approach to find the causes of accidents and breakdowns that have occurred as a result of certain factors and to take the necessary measures to prevent the same situations from happening again in the sector. The field of data mining, which is necessary to predict the future behavior of customers in the field of marketing, is an area that marketing also values. In this study, data mining studies to ensure safety in the aviation industry and the security of customer information in marketing will be emphasized, firstly, the concept and importance of data mining will be mentioned.

Data Mining in Customer Relationship Management in companies

More and more practice proves that focusing on customer needs is the key to business success.So studying on customer relationship management is very important for its implementation in enterprises.In recent years,data mining in customer relationship management(CRM) application has always been one of the hot spots.This paper shows the relevant methods of data mining application in CRM taking Telecom as an example.

Spatiotemporal deformation characteristics of Outang landslide and identification of triggering factors using data mining

Since the impoundment of Three Gorges Reservoir(TGR)in 2003,numerous slopes have experienced noticeable movement or destabilization owing to reservoir level changes and seasonal rainfall.One case is the Outang landslide,a large-scale and active landslide,on the south bank of the Yangtze River.The latest monitoring data and site investigations available are analyzed to establish spatial and temporal landslide deformation characteristics.Data mining technology,including the two-step clustering and Apriori algorithm,is then used to identify the dominant triggers of landslide movement.In the data mining process,the two-step clustering method clusters the candidate triggers and displacement rate into several groups,and the Apriori algorithm generates correlation criteria for the cause-and-effect.The analysis considers multiple locations of the landslide and incorporates two types of time scales:longterm deformation on a monthly basis and short-term deformation on a daily basis.This analysis shows that the deformations of the Outang landslide are driven by both rainfall and reservoir water while its deformation varies spatiotemporally mainly due to the difference in local responses to hydrological factors.The data mining results reveal different dominant triggering factors depending on the monitoring frequency:the monthly and bi-monthly cumulative rainfall control the monthly deformation,and the 10-d cumulative rainfall and the 5-d cumulative drop of water level in the reservoir dominate the daily deformation of the landslide.It is concluded that the spatiotemporal deformation pattern and data mining rules associated with precipitation and reservoir water level have the potential to be broadly implemented for improving landslide prevention and control in the dam reservoirs and other landslideprone areas.

A systematic study of Erzhu Erchen decoction against damp-heat internalized type 2 diabetes based on data mining and experimental verification

Background:Erzhu Erchen decoction(EZECD),which is based on Erchen decoction and enhanced with Atractylodes lancea and Atractylodes macrocephala,is widely used for the treatment of dampness and heat(The clinical manifestations of Western medicine include thirst,inability to drink more,diarrhea,yellow urine,red tongue,et al.)internalized disease.Nevertheless,the mechanism of EZECD on damp-heat internalized Type 2 diabetes(T2D)remains unknown.We employed data mining,pharmacology databases and experimental verification to study how EZECD treats damp-heat internalized T2D.Methods:The main compounds or genes of EZECD and damp-heat internalized T2D were obtained from the pharmacology databases.Succeeding,the overlapped targets of EZECD and damp-heat internalized T2D were performed by the Gene Ontology,kyoto encyclopedia of genes and genomes analysis.And the compound-disease targets-pathway network were constructed to obtain the hub compound.Moreover,the hub genes and core related pathways were mined with weighted gene co-expression network analysis based on Gene Expression Omnibus database,the capability of hub compound and genes was valid in AutoDock 1.5.7.Furthermore,and violin plot and gene set enrichment analysis were performed to explore the role of hub genes in damp-heat internalized T2D.Finally,the interactions of hub compound and genes were explored using Comparative Toxicogenomics Database and quantitative polymerase chain reaction.Results:First,herb-compounds-genes-disease network illustrated that the hub compound of EZECD for damp-heat internalized T2D could be quercetin.Consistently,the hub genes were CASP8,CCL2,and AHR according to weighted gene co-expression network analysis.Molecular docking showed that quercetin could bind with the hub genes.Further,gene set enrichment analysis and Gene Ontology represented that CASP8,or CCL2,is negatively involved in insulin secretion response to the TNF or lipopolysaccharide process,and AHR or CCL2 positively regulated lipid and atherosclerosis,and/or including NOD-like receptor signaling pathway,and TNF signaling pathway.Ultimately,the quantitative polymerase chain reaction and western blotting analysis showed that quercetin could down-regulated the mRNA and protein experssion of CASP8,CCL2,and AHR.It was consistent with the results in Comparative Toxicogenomics Database databases.Conclusion:These results demonstrated quercetin could inhibit the expression of CASP8,CCL2,AHR in damp-heat internalized T2D,which improves insulin secretion and inhibits lipid and atherosclerosis,as well as/or including NOD-like receptor signaling pathway,and TNF signaling pathway,suggesting that EZECD may be more effective to treat damp-heat internalized T2D.

Forecasting the Academic Performance by Leveraging Educational Data Mining

The study aims to recognize how efficiently Educational DataMining(EDM)integrates into Artificial Intelligence(AI)to develop skills for predicting students’performance.The study used a survey questionnaire and collected data from 300 undergraduate students of Al Neelain University.The first step’s initial population placements were created using Particle Swarm Optimization(PSO).Then,using adaptive feature space search,Educational Grey Wolf Optimization(EGWO)was employed to choose the optimal attribute combination.The second stage uses the SVMclassifier to forecast classification accuracy.Different classifiers were utilized to evaluate the performance of students.According to the results,it was revealed that AI could forecast the final grades of students with an accuracy rate of 97%on the test dataset.Furthermore,the present study showed that successful students could be selected by the Decision Tree model with an efficiency rate of 87.50%and could be categorized as having equal information ratio gain after the semester.While the random forest provided an accuracy of 28%.These findings indicate the higher accuracy rate in the results when these models were implemented on the data set which provides significantly accurate results as compared to a linear regression model with accuracy(12%).The study concluded that the methodology used in this study can prove to be helpful for students and teachers in upgrading academic performance,reducing chances of failure,and taking appropriate steps at the right time to raise the standards of education.The study also motivates academics to assess and discover EDM at several other universities.

Integrated data mining and network pharmacology to discover a novel traditional Chinese medicine prescription against diabetic retinopathy and reveal its mechanism

Background:Diabetic retinopathy(DR)is currently the leading cause of blindness in elderly individuals with diabetes.Traditional Chinese medicine(TCM)prescriptions have shown remarkable effectiveness for treating DR.This study aimed to screen a novel TCM prescription against DR from patents and elucidate its medication rule and molecular mechanism using data mining,network pharmacology,molecular docking and molecular dynamics(MD)simulation.Method:TCM prescriptions for treating DR was collected from patents and a novel TCM prescription was identified using data mining.Subsequently,the mechanism of the novel TCM prescription against DR was explored by constructing a network of core TCMs-core active ingredients-core targets-core pathways.Finally,molecular docking and MD simulation were employed to validate the findings from network pharmacology.Result:The TCMs of the collected prescriptions primarily possessed bitter and cold properties with heat-clearing and supplementing effects,attributed to the liver,lung and kidney channels.Notably,a novel TCM prescription for treating DR was identified,composed of Lycii Fructus,Chrysanthemi Flos,Astragali Radix and Angelicae Sinensis Radix.Twenty core active ingredients and ten core targets of the novel TCM prescription for treating DR were screened.Moreover,the novel TCM prescription played a crucial role for treating DR by inhibiting inflammatory response,oxidative stress,retinal pigment epithelium cell apoptosis and retinal neovascularization through various pathways,such as the AGE-RAGE signaling pathway in diabetic complications and the MAPK signaling pathway.Finally,molecular docking and MD simulation demonstrated that almost all core active ingredients exhibited satisfactory binding energies to core targets.Conclusions:This study identified a novel TCM prescription and unveiled its multi-component,multi-target and multi-pathway characteristics for treating DR.These findings provide a scientific basis and novel insights into the development of drugs for DR prevention and treatment.

Detection of Knowledge on Social Media Using Data Mining Techniques

In light of the rapid growth and development of social media, it has become the focus of interest in many different scientific fields. They seek to extract useful information from it, and this is called (knowledge), such as extracting information related to people’s behaviors and interactions to analyze feelings or understand the behavior of users or groups, and many others. This extracted knowledge has a very important role in decision-making, creating and improving marketing objectives and competitive advantage, monitoring events, whether political or economic, and development in all fields. Therefore, to extract this knowledge, we need to analyze the vast amount of data found within social media using the most popular data mining techniques and applications related to social media sites.

Construction and Practice of Teaching Evaluation System Based on Data Mining

The teaching quality evaluation system based on data mining technology can accurately and fairly identify the core driving factors to improve teaching quality.This method adopts the analysis of big data correlation rules,including data collection and processing preparation steps,builds the data warehouse of association rules,and then generates an educational quality evaluation framework using the principle of data mining.Based on this,this paper analyzes the construction design and method of the teaching evaluation system under data mining,hoping to provide help for the improvement of the teaching evaluation system and the improvement of teaching quality.

Exploring the medication pattern and mechanism of action of traditional Chinese medicine in treating polycystic ovary syndrome with kidney deficiency and blood stasis based on data mining and network pharmacology

Background:Using network pharmacology to explore the potential molecular mechanism of traditional Chinese medicine in treating polycystic ovary syndrome(PCOS)with kidney deficiency and blood stasis syndrome.Method:Collect the related literature materials of PCOS with kidney deficiency and blood stasis syndrome treated by traditional Chinese medicine in four databases in recent ten years,extract the information of prescriptions and complete the frequency analysis.Traditional Chinese Medicine Systems Pharmacology Database was used to screen out the effective components.Use Online Mendelian Inheritance in Man and other databases to screen PCOS disease targets.The intersection targets obtained by clustering prescription and PCOS disease targets were submitted to STRING database for protein-protein interaction network analysis,and Gene Ontology(GO)and Kyoto Encyclopedia of Genes and Genomes pathways were analysed by Metascape.Result:There are 155 kinds of traditional Chinese medicines used in the literature.The most commonly utilized ones are Cuscutae Semen,Angelicae Sinensis Radix,and Rehmanniae Radix Praeparata.The results of the cluster analysis indicated that the plants most commonly found throughout the prescription were Leonuri Herba,Lycopi Herba,Dipsaci Radix,etc.GO results show that biological processes include cell reaction to organic nitrogen compounds and cell reaction to nitrogen compounds.The functional display of GO molecule includes cytokine receptor binding,signal receptor regulator activity and so on.Kyoto Encyclopedia of Genes and Genomes results show that the possible mechanisms of action are cancer pathway,an endocrine resistance signal pathway.Conclusion:Through data mining,the cluster prescription for PCOS with kidney deficiency and blood stasis syndrome is Leonuri Herba,Lycopi Herba,Dipsaci Radix,etc.The network pharmacology research of cluster prescription shows that the main drug components for treating PCOS with kidney deficiency and blood stasis syndrome are quercetin,kaempferol,luteolin,tanshinone IIA,etc.,which act on PTGS2,NCOA2,and other targets,and treat PCOS with kidney deficiency and blood stasis syndrome through cancer and endocrine resistance.

How does a data strategy enable customer value?The case of FinTechs and traditional banks under the open finance framework

The coupling of data and digital innovation opens the way for new business in the financial services sector,where customers are placed at the centre of decisions and data can help to develop customer knowledge.To carry out our research,we adopted a multi-case study approach to explore how a data strategy is developed in the retail banking industry,together with its relationship with customer value,paying particular attention to the heterogeneity between traditional banks and financial technology companies(FinTechs).Two main points emerged from the study.Firstly,there are three possible approaches to Open Finance,which are mainly defined by their different corporate cultures,organisational configurations,technological architecture and data value.Secondly,it is not enough to be a FinTech to be best placed to exploit the market,as some traditional banks share the FinTechs’approach to Open Finance.Designing new tailored products,customising their prices and offering them over the right channels through targeted communication are all data-driven initiatives that stem from cross-or up-selling potential,core to the retail banking industry for turning a customer into a cash flow,thus enabling value to be created for customers.Our findings additionally revealed that there is a form of external information asymmetry between the customer and the bank,and that there is also an internal asymmetry between bank departments,as their visibility on information about the same customer may differ.

Spatial Data Mining to Support Environmental Management and Decision Making--A Case Study in Brazil

The growth of geo-technologies and the development of methods for spatial data collection have resulted in large spatial data repositories that require techniques for spatial information extraction, in order to transform raw data into useful previously unknown information. However, due to the high complexity of spatial data mining, the need for spatial relationship comprehension and its characteristics, efforts have been directed towards improving algorithms in order to provide an increase of performance and quality of results. Likewise, several issues have been addressed to spatial data mining, including environmental management, which is the focus of this paper. The main original contribution of this work is the demonstration of spatial data mining using a novel algorithm with a multi-relational approach that was applied to a database related to water resource from a certain region of S^o Paulo State, Brazil, and the discussion about obtained results. Some characteristics involving the location of water resources and the profile of who is administering the water exploration were discovered and discussed.

Signal classification method based on data mining formulti-mode radar

For the multi-mode radar working in the modern electronicbattlefield, different working states of one single radar areprone to being classified as multiple emitters when adoptingtraditional classification methods to process intercepted signals,which has a negative effect on signal classification. A classificationmethod based on spatial data mining is presented to address theabove challenge. Inspired by the idea of spatial data mining, theclassification method applies nuclear field to depicting the distributioninformation of pulse samples in feature space, and digs out thehidden cluster information by analyzing distribution characteristics.In addition, a membership-degree criterion to quantify the correlationamong all classes is established, which ensures classificationaccuracy of signal samples. Numerical experiments show that thepresented method can effectively prevent different working statesof multi-mode emitter from being classified as several emitters,and achieve higher classification accuracy.

Incorporating Domain Knowledge into Data Mining Process:An Ontology Based Framework

With the explosive growth of data available, there is an urgent need to develop continuous data mining which reduces manual interaction evidently. A novel model for data mining is proposed in evolving environment. First, some valid mining task schedules are generated, and then au tonomous and local mining are executed periodically, finally, previous results are merged and refined. The framework based on the model creates a communication mechanism to in corporate domain knowledge into continuous process through ontology service. The local and merge mining are transparent to the end user and heterogeneous data ,source by ontology. Experiments suggest that the framework should be useful in guiding the continuous mining process.

A Distributed Approach of Big Data Mining for Financial Fraud Detection in a Supply Chain

Supply Chain Finance(SCF)is important for improving the effectiveness of supply chain capital operations and reducing the overall management cost of a supply chain.In recent years,with the deep integration of supply chain and Internet,Big Data,Artificial Intelligence,Internet of Things,Blockchain,etc.,the efficiency of supply chain financial services can be greatly promoted through building more customized risk pricing models and conducting more rigorous investment decision-making processes.However,with the rapid development of new technologies,the SCF data has been massively increased and new financial fraud behaviors or patterns are becoming more covertly scattered among normal ones.The lack of enough capability to handle the big data volumes and mitigate the financial frauds may lead to huge losses in supply chains.In this article,a distributed approach of big data mining is proposed for financial fraud detection in a supply chain,which implements the distributed deep learning model of Convolutional Neural Network(CNN)on big data infrastructure of Apache Spark and Hadoop to speed up the processing of the large dataset in parallel and reduce the processing time significantly.By training and testing on the continually updated SCF dataset,the approach can intelligently and automatically classify the massive data samples and discover the fraudulent financing behaviors,so as to enhance the financial fraud detection with high precision and recall rates,and reduce the losses of frauds in a supply chain.

The Use of Data Mining Techniques in Rockburst Risk Assessment

Rockburst is an important phenomenon that has affected many deep underground mines around the world. An understanding of this phenomenon is relevant to the management of such events, which can lead to saving both costs and lives. Laboratory experiments are one way to obtain a deeper and better understanding of the mechanisms of rockburst. In a previous study by these authors, a database of rockburst laboratory tests was created; in addition, with the use of data mining （DM） techniques, models to predict rockburst maximum stress and rockburst risk indexes were developed. In this paper, we focus on the analysis of a database of in situ cases of rockburst in order to build influence diagrams, list the factors that interact in the occurrence of rockburst, and understand the relationships between these factors. The in situ rockburst database was further analyzed using different DM techniques ranging from artificial neural networks （ANNs） to naive Bayesian classifiers. The aim was to predict the type of rockburst-that is, the rockburst level-based on geologic and construction characteristics of the mine or tunnel. Conclusions are drawn at the end of the paper.

Hydraulic metal structure health diagnosis based on data mining technology

In conjunction with association rules for data mining, the connections between testing indices and strong and weak association rules were determined, and new derivative rules were obtained by further reasoning. Association rules were used to analyze correlation and check consistency between indices. This study shows that the judgment obtained by weak association rules or non-association rules is more accurate and more credible than that obtained by strong association rules. When the testing grades of two indices in the weak association rules are inconsistent, the testing grades of indices are more likely to be erroneous, and the mistakes are often caused by human factors. Clustering data mining technology was used to analyze the reliability of a diagnosis, or to perform health diagnosis directly. Analysis showed that the clustering results are related to the indices selected, and that if the indices selected are more significant, the characteristics of clustering results are also more significant, and the analysis or diagnosis is more credible. The indices and diagnosis analysis function produced by this study provide a necessary theoretical foundation and new ideas for the development of hydraulic metal structure health diagnosis technology.

A Distributed Data Mining System Based on Multi-agent Technology

Distributed Data Mining is expected to discover preciously unknown, implicit and valuable information from massive data set inherently distributed over a network. In recent years several approaches to distributed data mining have been developed, but only a few of them make use of intelligent agents. This paper provides the reason for applying Multi-Agent Technology in Distributed Data Mining and presents a Distributed Data Mining System based on Multi-Agent Technology that deals with heterogeneity in such environment. Based on the advantages of both the CS model and agent-based model, the system is being able to address the specific concern of increasing scalability and enhancing performance.