Customer relationship management data mining techniques naive bayes

customer relationship management data mining techniques naive bayes

Mining Technique. ಧಧNaive Keyword ü customer relationship management ; customer classification and prediction; data mining; naive Bayesian classifier. Customer relationship management based on data mining technique — Naive Bayesian classifier. Abstract: With the rampant competition in the domestic and. For Marketing, Sales, and Customer Relationship Management Gordon S. Linoff, Naïve Bayesian models provide a way out of this dilemma when you are.

Traditional mass marketing has focused on radio, television and newspapers as the media used to reach this broad audience. By reaching the largest audience possible, exposure to the product is maximized. There is an increasing awareness that effective customer relationship management can be done only based on an actual understanding of the needs and preferences of the customers. Under these condi- tions, data mining tools can help uncover the hidden infor- mation which resided already in the database.

But still there is a lack of such system which can target the customers according to their need. The existing system broad- casts the sales and promotion news to all that become more expensive and less effective.

Customers are important re- sources for an enterprise. Therefore, it is essential for, enter- prises to successfully acquire new customers and retain high value customers [6]. However, instead of targeting all customers equally or providing the same incentive offers to all customers, enterprises can select only those customers who meet certain profitability criteria based on their individual needs or purchasing behaviours [6].

These potential customers are the main contributor to the revenue of the company. Here main motto is to analyze the customer behavior and their purchasing activities so that a pattern can be ob- tained. For this purpose the data mining provides a technique for analysis and dependency analysis to discover the pattern and target the appropriate customer who can benefit the point of sales increment. Data mining techniques are deployed to scour large databases in order to find novel and useful patterns that might remain unknown.

The mining of gold from rocks or sand is referred to as gold mining rather than rock or sand mining. Data mining, also known as knowledge discovery in databases is a rapidly emerging. This technology is motivated by the need of new techniques to help analyze, understand or even visualize the huge amounts of stored data gathered from business and scientific applica- tions.

It is the process of discovering interesting knowledge, such as patterns, associations, changes, anomalies and signifi- cant structures from large amounts of data stored in data- bases, data warehouses, or other information repositories.

It can be used to help companies to make better decision to stay competitive in the marketplace. The major data mining func- tions that are developed in commercial and research commu- nities include summarization, association, classification, pre- diction and clustering.

These functions can be implemented using a variety of technologies, such as database-oriented techniques, machine learning and statistical techniques.

For the intended purpose we have used clustering and classifica- tion. Sequential pattern mining has become the challenging task in data mining due to complexity. Most common tools are statistics and set theory. Visualization refers the presentation of data so that users can view complex patterns. Visualization involves mapping of the data into some types of drawing or graphical objects. The visualization also helps in acquiring knowledge more comprehensively and most important, very quickly.

Data in a database can be viewed at different combinations of attributes. Applications of data mining for CRM The applications of data mining are divided into following various categories [ 3 ]. Some of the few applications domains and discussed how data mining tools should be developed for the such applications. Financial data collected in the banking and financial industry are often relatively complete, reliable and of high quality which facilities systematic data analysis and data mining.

Some examples cases such as Design and construction of data warehouses for multidimensional data analysis and data mining, Loan payment prediction and customer credit policy analysis, Classification and clustering of customers of targeting marketing, Detection of money laundering and other financial crimes.

It is the major application area of data mining because to identify customer buying behavior, achieve good customer retention. Retails data mining can help identify customer buying behaviors and achieve better customer retention and satisfaction. Some examples of retail industry are Design and construction of data warehouses based on the benefits of data mining,Multidimensional analysis of sales, customersproductstime and region, Analysis of the effectiveness of sales campaigns, Customer retention.

Health Care and Insurance Industry: Data mining also applied in claims analysis such as identify which medical procedures are claimed together. The techniques which are used for fraud detection in insurance. In insurance, four categories such as home insurance, life insurance, motor insurance, medical insurance.

Among the four, the motor and medical insurance have much more fraud problems. The data mining techniques which are more helpful for detecting the fraud in the insurance sector. Telecommunication market is rapidly expanding and highly competitive. It creates a demand for data mining in order to understand the business, identify the telecommunications patterns and improve the quality of service.

The scenarios for which data mining may improve telecommunications services as such as Multidimensional analysis of telecommunications data, Fraudulent pattern analysis and the identification of unusual patterns, Mobile telecommunications, use of visualization tools in telecommunications data analysis.

There was a problem providing the content you requested

Literature Survey Keshav Dahal et al. This classifier is so common and easy to implement and fast. The experiment were conducted using the various classification algorithms and compare the performance among then classifiers such as decision tree, neural network etc. He improved the lot of changes to enhance the performance and accuracy better. It considers for making classification useable is to identify a similar group of data from the whole training set of data and then training each group of similar data.

It has shown better results than the other classifier Decision tree. The experiment analysis was tested with the help of thyroid benchmark dataset. This achieves better results for a data mixing up with supervised and unsupervised learning Narender Kumar et al.

customer relationship management data mining techniques naive bayes

Cluster analysis K-means find the group of persons belongs which criteria. The customer data of LIC have taken for the experiment. Only the age and three premium policy are used for analysis. Cluster analysis using K-means to find the distance between the three customers. K-means is suitable technique for cluster analysis. It may set a path and make a good relationship between the customer and insurance policy organization. This method is to find the cluster C1 have the three customers S1,S2,S10 which satisfied with all the benefits terms and conditions of cluster same as the S1,S2,S10 then allocated the cluster C1.

Customer relationship management based on data mining technique — Naive Bayesian classifier

Cluster C2, C3 allocated as the cluster C1. It will increase the profit of the organization. Clustering optimization method is used to find the appropriate or local optimal solution.

Indranil Bose et al. It developed a model for the prediction of customer churn. The important decision in customer churn management is the separation of churners from non-churners. Decision tree model are very popular in prediction of churn. It used multiple variables for clustering and examines different approaches of hybridization for utilizing the results of clustering in order to build supervised learning models for prediction of churn.

In the hybrid method, clustering used as a first stage and decision tree used as a second stage. Three customers churn dataset used in this paper.

customer relationship management data mining techniques naive bayes

Yaya Xie et al. In this study, proposed a improved balanced random forests method IBRF.

customer relationship management data mining techniques naive bayes

The experimented were conducted with the help of real bank customer churn dataset. IBRF proved that better prediction results among the random forests such as balanced random forests and weighted random forests. The proposed method combines the two random forests such as balanced random forest and weighted random forest. This method to be proven that better accuracy, faster training speeds. It also examines the challenges of using data mining technology for predicting the customer behavior.

In this analysis, they have experimented on For experimental analysis, we eliminate some attributes because too many attributes used it is difficult to interpret. In this paper, posteriori classification process applied for the data. This technique helps us to increase the revenue of the organization Prabha Dhandayudam et al. Then the performance of the algorithm compared with other traditional techniques such as K-means, single link and complete link. RFM is very effective method for customer segmentation.

For segmenting the customers, the attribute R, F and M are used as three in clustering techniques.

Naïve Bayes Classifier - Fun and Easy Machine Learning

For finding the distance between from each object to all other object, here Manhattan distance used and store it in distance matrix. It experimented with real data set of the customer transaction details are used for clustering. In each iteration the pair of each cluster of same distance is merged in parallel instead of merging only one pair of cluster at a time. The parallel merging of clusters pairs improves the quality of clustering algorithm.

Customer classification in retail marketing by data mining

It will improve the performance the clustering algorithm better than the other traditional clustering algorithm. The performance of the clustering algorithms were measured in term of four criteria MSE, Intra cluster distance, Inter Cluster distance, Intra cluster distance divided by inter cluster distance. In this paper, the cluster technique used for customer segmentation. Isakki alias Devi et al. It used the Clustering and association rule find to identify customer behavior.

It can easily predict the sales. The customer with similar purchasing behavior are first grouped by means of clustering techniques such as K-means method and for each cluster an association rule Apriori algorithm to identify the products that are brought together by the customers. Association rules are adopted to discover the relationship and knowledge of the database.

Data analysis done by the open source data mining tool such as WEKA. Analysis of customer behavior aims to improve the overall performance of the enterprise. This paper focused on getting more customer satisfaction. Bart Baesens et al.

It improves marketing decision making. Bayesian network classifier used for customer life cycle slope estimation problem. They concluded that Bayesian network classifiers are performed well in predicting the future customer evolution. It augmented that loyal customers be always a regarded as a homogeneous group of profitable customers of a company, In this study.

Bayesian network classifiers have a good performance. This major focused the predictability of the sign of the slope and compare the performance of Bayesian network with other artificial intelligence technique. In this studythey tried to acknowledge the heterogeneity in the long-life customer and it is proved that possible to predict the slope of customer life cycle of long life customers.

customer relationship management data mining techniques naive bayes

To measure the performance of classifierthe PCC performance of correctly classified used. In this paper, clearly stated that bayesian network classifier is suitable for the customer lifecycle estimation problem and Markov blanket concept effective for attribute selection. Clustering algorithm-means problems occurs when empty clusters, 2.

Low comprehensibility used algorithms K-means 7. Bayesian Network Sometimes, it computationally 8. Classification, Semi supervised learning Cannot say too much in terms of The experimented were conducted using the database of the customers of the company dealing with selling of the vehicles.

For classification of customers using ART algorithm. The performance of this algorithm compared with back propagation algorithm. This algorithm taken only less time to provide the customer classification The time complexity of this algorithm is less than the backpropagation algorithm The algorithm was implemented in MATLAB 7. Siavash Emtiyaz et al. It is used to predict the category of an unknown customer. Semi- supervised learning SSL is a halfway between supervised and unsupervised learning.