Banking Credit Scoring Assessment Using Predictive K-Nearest Neighbour (PKNN) Classifier

Credit scoring models is a scientific methodology adopted by credit providers to assess the credit worthiness of applicants. The primary objective of such models has been to predict the potentiality of the loan applicant. A proper evaluation of the credit can help the service provider to determine whether to grant or to reject credit. Therefore, the objective of the study is to predict banking credit scoring assessment using Predictive K-Nearest Neighbour (PKNN) classifier. For the purpose of analysis two different credit approval datasets: Australian credit and German credit have been used. The results from the study show that the proposed model used for classification works better on credit dataset. Here, the study firstly attempted to find the optimal ‘K' value of the neighbourhood so that the classifier is tuned to forecast the credit worthiness and secondly, validated our proposed model on two credit approval datasets by checking the performance of the proposed models on the basis of classification accuracy.

Improved Seating Plans for Movie Theatre to Improve Revenue

Bollywood is not only one of the biggest film producers in India but also one of the largest centers of film production in the world. Seat occupancy rate and pricing of each seat are important parameters that determines the revenue of a cinema business. The objective of the chapter is to enable theater managers to determine the prices at the time of booking according to the occupancy rate so that the revenue is improved based on preferred demand for the respective seats. A multi criteria analysis is applied with seat occupancy rate as dependent variable and other factors as independent variables like Show time, Poster Size, Day of week and Timing of Release. Further, a predictive analysis can be carried out to determine the occupancy rate for the upcoming movies. Based on the occupancy rate, the managers at theater can adopt variable pricing concept to improve the revenue. This work shows an integrated method to develop a seating plan based on occupancy rate to improve the revenue using EMSR-b heuristic with an illustrated example for a theater.

Fuzzy Clustering

This chapter presents a study of the development of the clustering methodology to data analysis, with particular attention to the analysis from a crisp environment to a fuzzy environment. An applied problem concerning service quality (using SERVQUAL) of mobile phone users, and subsequent loyalty and satisfaction forms the data set to demonstrate the clustering issue. Following details on both the crisp k-means and fuzzy c-means clustering techniques, comparable results from their analysis are shown, on a subset of data, to enable both graphical and statistical elucidation. Fuzzy c-means is then employed on the full SERVQUAL dimensions, and the established results interpreted before tested on external variables, namely the level of loyalty and satisfaction across the different clusters established.

Prediction of the Quality of Fresh Water in a Basin

Derivatives play an important role in social and economic studies. They describe the behavior of conditional expectations. Once a phenomena is characterized by parametric specifications, the conditional expectation m(x) may be modeled by a regression function. Then, derivatives may be computed by fitting the regression function. In applications, parametric estimators are commonly used, because of the un-knowledge of other more effective methods. The validity of a regression fitting approach depends on the knowledge of certain aspects related with the true functional form. In this paper, we develop a study on the usage of soft computing methods for providing an alternative to the use of non-parametric regression. We develop our modeling including neural networks and rough sets approaches. The studied problem is the eutrophication due to the growth of the population of algae. Real life data is provided by a study on a fresh water basin. They are used for developing a comparison of different approaches. A methodology is recommended for implementing a monitoring system of the water quality.

Improving Customer Experience Using Sentiment Analysis in E-Commerce

In today's world, millions of online users post their opinions on product features, services, quality, benefits and other values of the products. These opinions or sentiment data generated via different communication mediums often include vital data points that can be fruitful for businesses in understanding customer experiences, products quality and services. The E-commerce companies considered social media platform for new product launch, promotion of products and features or establishing a successful business to customer relationship which produces great results. Analytics on this Social media data helps in identifying the customers in the right demographic, psychographic and lifestyle group. This chapter identifying important characteristics of customer reviews which help businesses houses to improve their marketing strategies.

Applications of the Stochastic Multicriteria Acceptability Analysis Method for Consumer Preference Study

Introducing a new product to the market is a complex, costly and time-consuming process which requires research on consumer preferences. On the basis of information on the characteristics of the new product and its competitors, as well as on the competitors and their market shares, the company has to estimate future market shares and to determine the profile of potential consumers inclined to purchase the new product. The purpose of our paper is to present a method of consumer preference research when introducing a new product, using a multiple criteria method called Stochastic Multicriteria Acceptability Analysis (SMAA). To apply this method, no information requiring tedious research is needed. SMAA allows to obtain essential information on the potential market power of the new product already at an early stage of its preparation. Furthermore, the flexibility of the SMAA method allows to easily expand the scope of the analysis by including additional information and various techniques of the modeling of the consumer selection process.

Operating Commodities Market by Automated Traders

This is an introductory work to the field of automatizing futures markets, related to commodities, so far operated by human traders. First, we build a mathematical framework for a futures market with many producers and consumers represented by automated traders in the market platform. Then we suggest an automatic trading strategy for the automatons. This strategy takes into account the forecasts of supply and demand streams as well as the evolution of nominal price. Later, we recall a set of analytical criteria used to measure the performance of a trading strategy. Next, we illustrate our approach by showing a price pattern generated by the automatic strategy and calculate its performances. Finally, we exhibit a heuristic based on simulation allowing to compute a quasi-optimal parameters matrix for this automatic trading system.

Car Safety

Car safety is an essential feature of marketing strategies for automobile companies. In this work, a statistical analysis on crash tests is conducted based on data available from European New Car Assessment Programme (Euro NCAP). The research work developed in this chapter presents a statistical analysis of the information produced by Euro NCAP, using the SPSS and MATLAB software, and seeks to answer the following research questions: - are there statistically significant differences on adult occupant safety in the six years under study? - are there statistically significant differences among the best-selling car classes regarding safety in frontal collisions? - are electric and hybrid automobiles less secure than their traditional counterparts with respect to frontal collisions?

E-Retailing from Past to Future

In this chapter, first, the definition, the advantages and disadvantages of e-retailing are given and the related literature about e-retailing is briefly explained in order to give a background to unfamiliar readers. Then, the qualitative and quantitative criteria which affect the e-retailer selection are determined and some e-retailers are evaluated by using a multi-phase, integrated Multi Criteria Decision Making (MCDM) approach. In first phase of proposed MCDM approach, the weight of each criterion is determined. In second phase, a pre-evaluation is made and some of the e-retailers are eliminated. In last phase, the remaining retailers are evaluated and the best one is determined. Finally, the study is concluded by discussions, inferences and recommendations for future work.

Fuzzy Multi-Objective Association Rule Mining Using Evolutionary Computation

In this chapter, we model association rule mining as a Fuzzy multi-objective global optimization problem by considering several measures of strength such as support, confidence, coverage, comprehensibility, leverage, interestingness, lift and conviction by utilizing various fuzzy aggregator operators. In this, pdel, each measure has its own level of significance. Three fuzzy multi-objective association rule mining techniques viz., Fuzzy Multi-objective Binary Particle Swarm Optimization based association rule miner (FMO-BPSO), a hybridized Fuzzy Multi-objective Binary Firefly Optimization and Threshold Accepting based association rule miner (FMO-BFFOTA), hybridized Fuzzy Multi-objective Binary Particle Swarm Optimization and Threshold Accepting based association rule miner (FMO-BPSOTA) have been proposed. These three algorithms have been tested on various datasets such as book, food, bank, grocery, click stream and bakery datasets along with three fuzzy aggregate operators. From these experiments, we can conclude that Fuzzy-And outperforms all the other operators.