Accepted Papers


  • A Study on the Causes for Failures in Mathematics by Engineering Students of Chennai Using Triangular Extended Fuzzy Clustering Model (TrEFCLM)
    A.Praveen Prakash , Esther Jerlin. J and BENNILO FERNANDES J,Hindustan University, India

    The aim of this paper is to introduce a new Fuzzy Model called Triangular Extended Fuzzy Clustering Model. In this paper the algorithm for this model is derived for the first time. Engineering students mostly keep arrear in engineering mathematics compared to other subjects. Hence Triangular Extended Fuzzy Clustering model is used to analyse the dominant causes for such failure to occur. In our survey, 100 engineering students were interviewed and their reasons for their failure in mathematics were taken as attributes and the above said model was applied to categorize the causes into three clusters namely Low, Medium and High. This paper consists of four sections. Section one gives the introduction of the problem and also the justification for having chosen to use the “Triangular Extended Fuzzy Clustering Modelae approach to obtain the dominant causes for the failure. Section two gives the preliminaries and the basics of Triangular Extended Fuzzy Clustering model. Section three deals with the application of the model in determining the cluster of problems, that fall under the three categories viz, aelowae, aemoderateae and aehighae . and the final fourth section gives the conclusion and suggestions based on the result.

  • Comparative Study of various training algorithms of Artificial Neural Networks on Diabetes dataset
    Sumi Alice Saji and Balachandran K,Christ University, India

    Artificial Neural Networks play an important role in diverse applications. They are a system of interconnected neurons that perform computation similar to a biological neural network. This study focuses on various training algorithms used in artificial neural networks. We have used the Pima Indian diabetes data set from UCI Machine Learning Repository for constructing and training the neural network. The system is implemented in MatlabR2013. About 768 instances were used for training the neural network. The input vector has patient history and the target output will be class label as tested positive or tested negative. Results show that out of the training algorithms, Levenberg-Marquardt Algorithm performed the best results. Even though, Resilient Backpropagation and Conjugate gradient with Powell/Beale Restarts Algorithms perform invariably, LM performs faster than other two algorithms.

  • A System to Detect Inappropriate Messages in Online Social Networks
    Rohan Shetty, Kalyani Nair, Shivani Singh, Shantanu Nakhare, Gopal Upadhye,Savitribai Phule Pune University, India

    As social networking is growing at a rapid pace today it is vital that we work on improving its management. Research has shown that the content present in online social networks may have significant influence on impressionable minds. If such platforms are misused, it will lead to negative consequences. Detecting insults or inappropriate messages continues to be one of the most challenging aspects of Online Social Networks (OSNs) today. We address this problem through a Machine Learning Based Soft Text Classifier approach using Support Vector Machine algorithm. The proposed system acts as a screening mechanism the alerts the user about such messages. The messages are classified according to their subject matter and each comment is labeled for the presence of profanity and insults.

  • Fuzzy Techniques for Transportation Engineering: The State of the Art
    Marisamynathan S and Vedagiri P,Indian Institute of Technology Bombay, India

    This paper presents a classification and analysis of the results achieved using fuzzy for transportation engineering problems. The basic fundamentals of fuzzy set theory, fuzzy logic and fuzzy clustering are presented as well as applications of fuzzy to solve traffic and transportation engineering problems. The basic premise of fuzzy regression model are presented by developing two steps fuzzy regression model for fuzzy dependent variable and one crisp independent variable. Fuzzy regression model is proposed to develop Level of Service (LOS) model and fuzzy clustering is suggested to fix criteria for LOS boundary at signalized intersections. Finally the proposed new methodology is developed and applied with numerical example.