Laxman Sahoo

Novel Outlier Detection by Integration of Clustering and Classification

A unique method of outlier detection consisting of integration of clustering and classification is proposed here. Basically the algorithm is divided into two parts the first phase consists of application of the classical DBSCAN algorithm to the data set which is followed by the second phase which consists of application of decision tree classification algorithm. The analysis on the algorithm states that the accuracy of unwanted data detection is high in the proposed method.

Recognition of Telecom Customer’s Behavior as Data Product in CRM Big Data Environment

This paper approaches toward the standardization of telecom customer’s behavior by specifying the call activities like frequency, duration, time of calls with the type of calls like local, national, and international. In the same way specifying the SMS/MMS activities as behavior of customers plus the rate of data pack and talk-time recharge. It is an attempt to identify meaningful attributes to describe behavior of customer plus study the available call detail records in big data environment and recognize the procedure that uses customer behavior for the designing of data product which is tariff plan.

Architecture for the Strategy-Planning Techniques Using Big Data Analytics

The rapid growth in technology and market has posed a throat-cut competition among the service providers. Retaining of existing customers in place of catching new one is 90% cheaper, but it needs to know the customer very well, which is possible by analyzing the customer data. To analyze customer data and provide customer-oriented services, this paper recommends an architecture for the development of techniques which would further be able to design strategies, such as tariff plan, on the basis of available information from the customer relationship management system of any service-based company, which is finally known as customer-oriented data product. This architecture has five phases with data source, data collection, data refinement, analysis of collected data, and generation of data product. This paper summarizes the requirement of data sets and systems to develop strategy-planning techniques with the study of available architectures in the big data and CRM environment.

Development of Policy Designing Technique by Analyzing Customer Behavior Through Big Data Analytics

Technological developments and market trends are two leading affairs of the current era, posing customer as most important entity to be caught. Use of big data analytics to retain customers by offering them customer-oriented policies and making them feel important and precious for the service-providing company is the core thought behind this research paper. A framework to obtain process and analyze service usage data, with a new algorithm known as Altered Genetic K-Means clustering algorithm based on mapReduce is presented here. This paper implements mapReduce-based Altered Genetic K-Means Clustering (AGKM) algorithm on data acquired from BSS/OSS of telecom CRM and cleaned by R, to categorize customers having similar call activities. Results show that specific group of customers such as students, senior citizens, housewives, business people, and employees can be identified and according to their call timings, durations, call types, net usage, etc., policies (tariff plans in this case) can be designed. The novelty of this work is in its thought of capturing customers by knowing them well in place of first predicting churn and then taking action.

Personalized Movie Recommendation System Using Twitter Data

Nowadays, we are living in an age recommendation, but the proper recommendation needs more accurate and relevant datas as their inputs. Rating databases like MovieLence or Netflix have long been popular and being widely used in recommendation system areas for research in past decades. But nowadays, they become irrelevant due to lack of new and relevant datas. Nowadays, social media like Facebook and Twitter become the most popular for researchers due to availability of large amount of new and relevant datas. In this paper, we have built a recommendation engine by analyzing rating datasets collected from Twitter to recommend movies to specific user using R.

AKM—Augmentation of K-Means Clustering Algorithm for Big Data

Clustering for big data analytics is a growing subject due to the large size of variety data sets needed to be analyzed in distributed and parallel environment. An augmentation of K-Means clustering algorithm is projected and evaluated here for MapReduce framework by using the concepts of genetic algorithm steps. Chromosome formation, fitness calculation, optimization, and crossover logics are used to overcome the problem of suboptimal solutions of K-Means clustering algorithm and reduction of time complexity of genetic K-Means algorithm for big data. Proposed algorithm is not dealing with the selection of parents to be sent to mating pool and mutation steps, so the performance time is improved.

Survey on CRM Analytics in Telecommunication Industry

The goal of data analytics is uncovering fruitful information and decision making from very large data sets in respect to various applications, which can redefine and improve customer relationships. With the increasing need of customizing every product and services the concept of Customer Relationship Management has been introduced. CRM is practiced in business by detecting patterns in customer data. Telecommunication sector is most well known and important to be dealing with customer interests than anyone else since they are well aware of their customers and can efficiently keep record of their customers’ actions. As the colossal volume of data produced by telecommunication companies cannot be evaluated manually, varied data analytics approaches are needed to be applied. Data analytics assists a business grasp its customers in a much superior way. In this paper we will make a survey on the concept of CRM and the use of data analytics in it.

Design and implementation of Three Phase Commit Protocol (3PC) algorithm

Transaction Management plays the key role in any type of database system. In order to make the Application Query, many protocols have been designed. In this paper, we have developed very efficient and enhanced the Three Phase Commit Protocol (3PC), after studying the drawback of 2PC. In 2PC, we observed that if two side transaction will perform in 2PC(i.e., one side called sender and other side called a receiver). When sender side communicates with the receiver side in the form of performing initial values prepare for commit or abort message. They have the some possibilities both sides that are, case 1- if the sender side sends the commit data to receiver side that may commit both sides. Case 2- if the sender side sends the data but data will abort to the sender side and the receiver side also data will abort. Case 3- if the sender side sends the commit data to receiver side but receiver side data will abort. Case 4- if the sender side sends the commit data but receiver side, will not ensure that at a time data will commit or abort. Overall, according to all possibilities, we have four cases which are applied. We investigated that data is not sure from both side (sender side and receiver side). In this paper we also observed that if 3PC will use such type of cases then it will avoid blocking problems, because after abort/failed the data of 2PC protocol. Data are blocked and reduced the blocking problem through 3PC techniques. It has one active data for backup, if failed/abort the data to both sides. It has stored and adds multiple sites for decision pays for committing not for abort. It is called pre-commit decision process and record of the data are stored in multiple sites (i.e., K sites), and we also implemented 3PC Algorithm in this paper.