Papers by Arindam Chaudhuri
arXiv (Cornell University), Jul 7, 2013
In India financial markets have existed for many years. A functionally accented, diverse, efficie... more In India financial markets have existed for many years. A functionally accented, diverse, efficient and flexible financial system is vital to the national objective of creating a market-driven, productive and competitive economy. Today markets of varying maturity exist in equity, debt, commodities and foreign exchange. Of the 25 stock markets in the country, the most important is Bombay Stock Exchange. In this work we attempt to generate prediction rules scheme for stock price movement at Bombay Stock Exchange using an important Soft Computing paradigm viz., Rough Fuzzy Multi Layer Perception. The use of Computational Intelligence Systems such as Neural Networks, Fuzzy Sets, Genetic Algorithms, etc. for Stock Market Predictions has been widely established. The process is to extract knowledge in the form of rules from daily stock movements. These rules can then be used to guide investors. To increase the efficiency of the prediction process, Rough Sets is used to discretize the data. The methodology uses a Genetic Algorithm to obtain a structured network suitable for both classification and rule extraction. The modular concept, based on divide and conquer strategy, provides accelerated training and a compact network suitable for generating a minimum number of rules with high certainty values. The concept of variable mutation operator is introduced for preserving the localized structure of the constituting Knowledge Based sub-networks, while they are integrated and evolved. Rough Set Dependency Rules are generated directly from the real valued attribute table containing Fuzzy membership values. The paradigm is thus used to develop a rule extraction algorithm. The extracted rules are compared with some of the related rule extraction techniques on the basis of some quantitative performance indices. The proposed methodology extracts rules which are less in number, are accurate, have high certainty factor and have low confusion with less computation time. To the best of our knowledge, this is the first Rough Fuzzy Multi Layer Perception method for discovering stock prediction rules of Bombay Stock Exchange.
arXiv (Cornell University), Mar 13, 2015
Mixed Integer Optimization has been a topic of active research in past decades. It has been used ... more Mixed Integer Optimization has been a topic of active research in past decades. It has been used to solve Statistical problems of classification and regression involving massive data. However, there is an inherent degree of vagueness present in huge real life data. This impreciseness is handled by Fuzzy Sets. In this Paper, Fuzzy Mixed Integer Optimization Method (FMIOM) is used to find solution to Regression problem. The methodology exploits discrete character of problem. In this way large scale problems are solved within practical limits. The data points are separated into different polyhedral regions and each region has its own distinct regression coefficients. In this attempt, an attention is drawn to Statistics and Data Mining community that Integer Optimization can be significantly used to revisit different Statistical problems. Computational experimentations with generated and real data sets show that FMIOM is comparable to and often outperforms current leading methods. The results illustrate potential for significant impact of Fuzzy Integer Optimization methods on Computational Statistics and Data Mining.
arXiv (Cornell University), Mar 13, 2015
Multiple Air Vehicles (AVs) to prosecute geographically dispersed targets is an important optimiz... more Multiple Air Vehicles (AVs) to prosecute geographically dispersed targets is an important optimization problem. Associated multiple tasks viz., target classification, attack and verification are successively performed on each target. The optimal minimum time performance of these tasks requires cooperation among vehicles such that critical time constraints are satisfied i.e. target must be classified before it can be attacked and AV is sent to target area to verify its destruction after target has been attacked. Here, optimal task scheduling problem from Indian Air Force is formulated as Fuzzy Mixed Integer Linear Programming (FMILP) problem. The solution assigns all tasks to vehicles and performs scheduling in an optimal manner including scheduled staged departure times. Coupled tasks involving time and task order constraints are addressed. When AVs have sufficient endurance, existence of optimal solution is guaranteed. The solution developed can serve as an effective heuristic for different categories of AV optimization problems.
arXiv (Cornell University), Jul 7, 2013
In India financial markets have existed for many years. A functionally accented, diverse, efficie... more In India financial markets have existed for many years. A functionally accented, diverse, efficient and flexible financial system is vital to the national objective of creating a market-driven, productive and competitive economy. Today markets of varying maturity exist in equity, debt, commodities and foreign exchange. Of the 25 stock markets in the country, the most important is Bombay Stock Exchange. In this work we attempt to generate prediction rules scheme for stock price movement at Bombay Stock Exchange using an important Soft Computing paradigm viz., Rough Fuzzy Multi Layer Perception. The use of Computational Intelligence Systems such as Neural Networks, Fuzzy Sets, Genetic Algorithms, etc. for Stock Market Predictions has been widely established. The process is to extract knowledge in the form of rules from daily stock movements. These rules can then be used to guide investors. To increase the efficiency of the prediction process, Rough Sets is used to discretize the data. The methodology uses a Genetic Algorithm to obtain a structured network suitable for both classification and rule extraction. The modular concept, based on divide and conquer strategy, provides accelerated training and a compact network suitable for generating a minimum number of rules with high certainty values. The concept of variable mutation operator is introduced for preserving the localized structure of the constituting Knowledge Based sub-networks, while they are integrated and evolved. Rough Set Dependency Rules are generated directly from the real valued attribute table containing Fuzzy membership values. The paradigm is thus used to develop a rule extraction algorithm. The extracted rules are compared with some of the related rule extraction techniques on the basis of some quantitative performance indices. The proposed methodology extracts rules which are less in number, are accurate, have high certainty factor and have low confusion with less computation time. To the best of our knowledge, this is the first Rough Fuzzy Multi Layer Perception method for discovering stock prediction rules of Bombay Stock Exchange.
African Journal of Mathematics and Computer science Research, Feb 28, 2011
Research domain. The problem has been investigated under different headings and solved with vario... more Research domain. The problem has been investigated under different headings and solved with various approaches including soft computing and linear programming. The conventional linear programming deals with crisp parameters. However, information available in real life system is of vague, imprecise and uncertain nature. The impreciseness and uncertainty aspects are handled using Fuzzy Sets to obtain optimal solutions. Multi-Objective Linear Programming effectively deals with flexible aspiration levels or goals. Fuzzy Multi-Objective Linear Programming enhances the effectiveness of solutions with acceptable solutions through fuzzy constraints. In this work, Fuzzy Multi-Objective Linear Programming is used for solving TSP with vague and imprecise parameters. An example of TSP with multiple objectives and imprecise parameters is also discussed.
Fuzzy multi-objective linear programming for traveling salesman problem
arXiv (Cornell University), Jun 24, 2018
Connected vehicle fleets are deployed worldwide in several industrial IoT scenarios. With the gra... more Connected vehicle fleets are deployed worldwide in several industrial IoT scenarios. With the gradual increase of machines being controlled and managed through networked smart devices, the predictive maintenance potential grows rapidly. Predictive maintenance has the potential of optimizing uptime as well as performance such that time and labor associated with inspections and preventive maintenance are reduced. In order to understand the trends of vehicle faults with respect to important vehicle attributes viz mileage, age, vehicle type etc this problem is addressed through hierarchical modified fuzzy support vector machine (HMFSVM). The proposed method is compared with other commonly used approaches like logistic regression, random forests and support vector machines. This helps better implementation of telematics data to ensure preventative management as part of the desired solution. The superiority of the proposed method is highlighted through several experimental results.
Optical Character Recognition Systems for English Language
Studies in fuzziness and soft computing, Dec 24, 2016
The optical character recognition (OCR) systems for English language were the most primitive ones... more The optical character recognition (OCR) systems for English language were the most primitive ones and occupy a significant place in pattern recognition. The English language OCR systems have been used successfully in a wide array of commercial applications. The different challenges involved in the OCR systems for English language is investigated in this chapter. The pre-processing activities such as binarization, noise removal, skew detection and correction, character segmentation and thinning are performed on the datasets considered. The feature extraction is performed through discrete cosine transformation. The feature based classification is performed through important soft computing techniques viz fuzzy multilayer perceptron (FMLP), rough fuzzy multilayer perceptron (RFMLP), fuzzy support vector machine (FSVM) and fuzzy rough support vector machine (FRSVM). The superiority of soft computing techniques is demonstrated through the experimental results.
Probabilistic View of Operational Risk
Studies in fuzziness and soft computing, Oct 30, 2015
In this chapter we present the probabilistic view of operational risk which represented through g... more In this chapter we present the probabilistic view of operational risk which represented through g-and-h distribution. The concept of value at risk (VaR) is presented alongwith subadditivity of VaR. The subjective value at risk (SVaR) is proposed. The risk and deviation measures are also discussed. The equivalence of chance and VaR constraints are highlighted. Some advanced properties of g-and-h distribution are discussed. Some applications on probabilistic view of operational risk are given and then the decomposition is performed in accordance with contribution to risk factors.
QoS prediction for network data traffic using hierarchical modified regularized least squares rough support vector regression
In this paper, we present a novel approach for predicting QoS on networks having complex traffic ... more In this paper, we present a novel approach for predicting QoS on networks having complex traffic and multitenant architecture. We propose a soft computing based hierarchical modified regularized least squares rough support vector regression approach on network traffic to deliver better QoS. QoS prediction takes advantage of past service usage experiences. It does not consume any additional invocations of network services and avoids time consuming real world service incantations. We discuss the proposed approach and provide important aspects of QoS prediction. The experiments are conducted on real world public dataset and compared with benchmark dataset. The results show that proposed approach achieves high prediction accuracies than other techniques.
Optical Character Recognition Systems
Studies in fuzziness and soft computing, Dec 24, 2016
Optical character recognition (OCR) is process of classification of optical patterns contained in... more Optical character recognition (OCR) is process of classification of optical patterns contained in a digital image. The character recognition is achieved through segmentation, feature extraction and classification. This chapter presents the basic ideas of OCR needed for a better understanding of the book. The chapter starts with a brief background and history of OCR systems. Then the different techniques of OCR systems such as optical scanning, location segmentation, pre-processing, segmentation, representation, feature extraction, training and recognition and post-processing. The different applications of OCR systems are highlighted next followed by the current status of the OCR systems. Finally, the future of the OCR systems is presented.
Possibilistic View of Operational Risk
Studies in fuzziness and soft computing, Oct 30, 2015
In this Chapter the possibilistic view of operational risk is presented. In view of this the prob... more In this Chapter the possibilistic view of operational risk is presented. In view of this the probabilistic treatment of operational risk in Chap. 4 has been remodelled with possibility theory. The g-and-h distribution has been redefined as fuzzy g-and-h distribution. Similarly value at risk (VaR) concept is extended to fuzzy value at risk (VaR) and subadditivity of fuzzy VaR also defined. Based on fuzzy VaR the fuzzy subjective value at risk (SVaR) is highlighted. The risk and deviation measures are also extended to their fuzzy versions. An application of fuzzy SVaR optimization is also illustrated.
Optical Character Recognition Systems for Hindi Language
Studies in fuzziness and soft computing, Dec 24, 2016
arXiv (Cornell University), Jul 7, 2013
Transportation Problem is an important problem which has been widely studied in Operations Resear... more Transportation Problem is an important problem which has been widely studied in Operations Research domain. It has been often used to simulate different real life problems. In particular, application of this Problem in NP-Hard Problems has a remarkable significance. In this Paper, we present the closed, bounded and non-empty feasible region of the transportation problem using fuzzy trapezoidal numbers which ensures the existence of an optimal solution to the balanced transportation problem. The multi-valued nature of Fuzzy Sets allows handling of uncertainty and vagueness involved in the cost values of each cells in the transportation table. For finding the initial solution of the transportation problem we use the Fuzzy Vogel's Approximation Method and for determining the optimality of the obtained solution Fuzzy Modified Distribution Method is used. The fuzzification of the cost of the transportation problem is discussed with the help of a numerical example. Finally, we discuss the computational complexity involved in the problem. To the best of our knowledge, this is the first work on obtaining the solution of the transportation problem using fuzzy trapezoidal numbers.
arXiv (Cornell University), Apr 26, 2013
The Fuzzy Modeling has been applied in a wide variety of fields such as Engineering and Managemen... more The Fuzzy Modeling has been applied in a wide variety of fields such as Engineering and Management Sciences and Social Sciences to solve a number Decision Making Problems which involve impreciseness, uncertainty and vagueness in data. In particular, applications of this Modeling technique in Decision Making Problems have remarkable significance. These problems have been tackled using various theories such as Probability theory, Fuzzy Set Theory, Rough Set Theory, Vague Set Theory, Approximate Reasoning Theory etc. which lack in parameterization of the tools due to which they could not be applied successfully to such problems. The concept of Soft Set has a promising potential for giving an optimal solution for these problems. With the motivation of this new concept, in this paper we define the concepts of Soft Relation and Fuzzy Soft Relation and then apply them to solve a number of Decision Making Problems. The advantages of Fuzzy Soft Relation compared to other paradigms are discussed. To the best of our knowledge this is the first work on the application of Fuzzy Soft Relation to the Decision Making Problems.
arXiv (Cornell University), Jul 7, 2013
Examination Timetable Problem (ETP) is NP-Hard combinatorial optimization problem. It has receive... more Examination Timetable Problem (ETP) is NP-Hard combinatorial optimization problem. It has received tremendous research attention during the past few years given its wide use in universities. ETP can be defined as assignment of courses to be examined, candidates to time periods and examination rooms while satisfying a set of constraints which may be either hard or soft. Several methods have been proposed most of which are based on heuristics like Search techniques, Evolutionary Computation etc. In this Paper, we develop three mathematical models for Netaji Subhas Open University, Kolkata, India using Fuzzy Integer Linear Programming (FILP) technique. In most real life situations, information available in is not exact, lacks precision and has an inherent degree of vagueness. To deal with this we model various allocation variables through fuzzy numbers expressing lack of precision the decision maker has. The solution to the problem is obtained using Fuzzy number ranking method. Each feasible solution has fuzzy number obtained by Fuzzy objective function. The different FILP technique performance are demonstrated by experimental data generated through extensive simulation from Netaji Subhas Open University, Kolkata, India in terms of its execution times. The proposed FILP models are compared with commonly used heuristic viz. Integer Linear Programming approach on experimental data which gives an idea about quality of heuristic. The techniques are also compared with different Artificial Intelligence based heuristics for ETP with respect to best and mean cost as well as execution time measures on Carter benchmark datasets to illustrate its effectiveness. FILP paradigm takes an appreciable amount of time to generate satisfactory solution in comparison to other heuristics. The formulation thus serves as good benchmark for other heuristics. The experimental study presented here focuses on producing a methodology that generalizes well over spectrum of techniques that generates significant results for one or more datasets. The performance of FILP model is finally compared to the best results cited in literature for Carter benchmarks to assess its potential. The problem can be further reduced by formulating with lesser number of allocation variables it without affecting optimality of solution obtained. FLIP model for ETP can also be adapted to solve other ETP as well as combinatorial optimization problems. To the best of our knowledge this is first work on ETP using FILP technique.
InTech eBooks, Nov 30, 2010
Traveling Salesman Problem, Theory and Applications 198 instances as large as 24,978 cities to op... more Traveling Salesman Problem, Theory and Applications 198 instances as large as 24,978 cities to optimality. Trying to solve the course of exponentials parallel implementations of TSP were realized (Christof T. & Reinelt G., 1995). However, for practicability reasons specifically for large numbers of cities, heuristic approaches for solving TSP are very popular, which try to produce an optimal or close to optimal solution. It arises as sub-problem in many transportation and logistic applications (Chaudhuri A., 2007), for example the problem of arranging school bus routes to pick up children in a district. This application is of important significance to TSP since it provides motivation for Merrill Flood one of the pioneers of TSP research in 1940s. A second application from 1940s involved transportation of farming equipment from one location to another leading to mathematical studies by P. C. Mahalanobis and R. J. Jessen. More recent applications involve scheduling of service call at cable firms, delivery of meals to homebound persons, scheduling of stacker cranes in warehouses, routing of trucks for parcel post pickup etc. Although transportation applications are most natural setting for TSP, simplicity of the model has led to many interesting applications in other areas. A classic example is scheduling of machine to drill holes in circuit boards where holes to be drilled are cities and cost of travel is the time it takes to move the drill head from one hole to next. TSP has some direct importance, since quite a lot of practical applications can be put in this form. It also has theoretical significance in Complexity Theory (Garey M. & Johnson D., 1990) since TSP is one of the classes of NP-Complete Combinatorial Optimization Problems (Korte B. H. & Vygen J., 2008) which are difficult optimization problems where the set of feasible solutions or trial solutions which satisfy constraints of problem but are not necessarily optimal is finite, though usually very large set. The numbers of feasible solutions grow as some combinatorics factor such as ! n where, n characterizes size of the problem. It has often been the case that progress on TSP (Laporte G., 2010) has led to the progress on many Combinatorial Optimization Problems. In this way, TSP is an ideal stepping stone for study of Combinatorial Optimization Problems. Although many optimal algorithms exist for solving TSP it has been realized that it is computationally infeasible to obtain optimal solution to the problem. For large-size problem (Cormen T. H. et al., 2001) it has been proved that it is almost impossible to generate an optimal solution within reasonable amount of time. Heuristics instead of optimal algorithms are thus extensively used to solve such problems (Hansen M. P., 2000). Many heuristic algorithms give near optimal solutions to the problem which are used for practicability reasons specifically for large numbers of cities. Heuristic approaches (Lin S. & Kernighan B. W., 1973) for solving TSP are thus very popular which try to produce an optimal or close to optimal solution. The commonly used heuristic approaches are: (a) Greedy Algorithms; (b) 2-opt Algorithm; (c) 3-opt Algorithm; (d) Simulated Annealing; (e) Genetic Algorithms and (e) Artificial Neural Network (ANN). However, efficiencies vary from case to case and from size to size. Generally the most common heuristic is ANN which are well suited for solving problems that are hard to catch in mathematical models. However, the usage and employment of ANN in such application domains is often dependent on tractability of processing costs. The problem domains for employment of ANN are increasing (Haykin S., 2008) and also problem themselves are getting larger and more complex (Arbib M., 2003). This leads to larger networks consisting of huge numbers of nodes and interconnection links which results in exceeding costs for network specific operations such as evaluation and training. Especially the cost intensive training phase of ANN inherits a major drawback due to the www.intechopen.com A Study of Traveling Salesman Problem Using Fuzzy Self Organizing Map
arXiv (Cornell University), Jul 7, 2013
Transportation Problem is an important aspect which has been widely studied in Operations Researc... more Transportation Problem is an important aspect which has been widely studied in Operations Research domain. It has been studied to simulate different real life problems. In particular, application of this Problem in NP-Hard Problems has a remarkable significance. In this Paper, we present a comparative study of Transportation Problem through Probabilistic and Fuzzy Uncertainties. Fuzzy Logic is a computational paradigm that generalizes classical two-valued logic for reasoning under uncertainty. In order to achieve this, the notation of membership in a set needs to become a matter of degree. By doing this we accomplish two things viz., (i) ease of describing human knowledge involving vague concepts and (ii) enhanced ability to develop cost-effective solution to real-world problem. The multi-valued nature of Fuzzy Sets allows handling uncertain and vague information. It is a model-less approach and a clever disguise of Probability Theory. We give comparative simulation results of both approaches and discuss the Computational Complexity. To the best of our knowledge, this is the first work on comparative study of Transportation Problem using Probabilistic and Fuzzy Uncertainties.
Datasets Used
Springer eBooks, 2017
Uploads
Papers by Arindam Chaudhuri