Vellore Institute of Technology (VIT) is a private university located in&nbsp;Tamil Nadu, India. Founded in 1984, as Vellore Engineering College, the institution offers 20 undergraduate, 34 postgraduate, four integrated and four research programs. It has campuses in Vellore, Amravati, Bhopal and Chennai.

VIT is one of the top ranked private universities in India according to NIRF, THE and QS Rankings.&nbsp;Govt. of India has recognized&nbsp;VIT, Vellore as an&nbsp;Institution of Eminence. This has allowed VIT to take independent quality initiatives and move up in world ranking.

&nbsp;

&nbsp;

VIT University

Purpose: In today's world, the recommender systems are very valuable systems for the online users, as the World Wide Web is loaded with plenty of available information causing the online users to spend more time and money. The recommender systems suggest some possible and relevant recommendation to the online users by applying the recommendation filtering techniques to the available source of information. The recommendation filtering techniques take the input data denoted as the matrix representation which is generally very sparse and high dimensional data in nature. Hence, the sparse data matrix is completed by filling the unknown or missing entries by using many matrix completion techniques. One of the most popular techniques used is the matrix factorization (MF) which aims to decompose the sparse data matrix into two new and small dimensional data matrix and whose dot product completes the matrix by filling the logical values. However, the MF technique failed to retain the loss of original information when it tried to decompose the matrix and the error rate is relatively high which clearly shows the loss of such valuable information. Design/methodology/approach: To alleviate the problem of data loss and data sparsity, the new algorithm from formal concept analysis (FCA), a mathematical model, is proposed for matrix completion which aims at filling the unknown or missing entries without loss of valuable information to a greater extent. The proposed matrix completion algorithm uses the clustering technique where the users who have commonly rated the items and have not commonly rated the items are captured into two classes. The matrix completion algorithm fills the mean cluster value of the unknown entries which well completes the matrix without actually decomposing the matrix. Findings: The experiment was conducted on the available public data set, MovieLens, whose result shows the prediction error rate is minimal and the comparison with the existing algorithms is also studied. Thus, the application of FCA in recommender systems proves minimum or no data loss and improvement in the prediction accuracy of rating score. Social implications: The proposed matrix completion algorithm using FCA performs good recommendation which will be more useful for today's online users in making decision with regard to the online purchasing of products. Originality/value: This paper presents the new technique of matrix completion adopting the vital properties from FCA which is applied in the recommender systems. Hence, the proposed algorithm performs well when compared to other existing algorithms in terms of prediction accuracy.

Department of Smart Computing

School of Information Technology and Engineering

Vellore Campus

World University root node

International Journal of Pervasive Computing and Communications

Three-way formal concept clustering technique for matrix completion in recommender system

Big data refers to the huge torrent of large-scale datasets that are being generated at an exponential growth. Since we live in this digital world, the era of big data has emerged in part and parcel of our lives. The emergence of big data has reached in almost several domains like healthcare industry, telecom industry, molecular biology, biochemistry, physics, astronomy, computer science, business and others. In this paper, we have termed the types of big data by the form SMSS data which is simply meaning social, mobile, spatial and sensor data. This paper aims to provide the importance of big data analytics brought over the different types of big data extracted from heterogeneous data sources. To achieve this objective, we have made an intensive study of several literatures and considered a variety of big data applications which are being discussed to showcase its value. Also, a generic framework is proposed that can be applicable to any kind of big data types extracted from such a diverse heterogeneous data sources. Finally, a few open source tools that can be used for processing the big data are presented.

VIT Unitversity

International Journal of Intelligent Enterprise

SMSS: does social, mobile, spatial and sensor data have high impact on big data analytics

In this digital world, the Internet is the major source of rich amount of data where there are large numbers of choices available to the internet users. It requires the attention to scrutinize, itemize and effectively project the essential information in order to violate the dispute of information overload. The information overload is an obstacle for many of the internet users. Recommender systems aim to provide solution to such obstacles by refining and seeking the large volume of exponentially growing information to provide the internet users with the personalized content and services. Here finding or predicting the user-item rating with the sparse data available in the rating matrix of the input data becomes a challenging task. This paper explores the problem of recommendation by presenting a FCA-based methodology which uses the mathematical tool in reducing the dimensionality of the user-item rating thereby determining the rating value of the unknown user with its corresponding item. The proposed method is simple and faster explained with an illustration to complete the user-item rating matrix from which the rating can be predicted by applying any of the recommendation algorithms.

Advances in Intelligent Systems and Computing

A New Approach for Matrix Completion Using Arrow Relation of FCA in Recommender Systems

Big data is a late of huge information stored in it and all we need is to dig into get the important information out of it and create a useful system which can be very helpful in improving the current scenario. There are various applications where big data is being used and even there are few fields that are learning techniques to go with big data and evaluate their work and get an improve decision. This paper particularly concentrates on the e commerce system which is highly trending on the market field. [20] E commerce also known as electronic commerce is a market place which gives you a platform to enjoy various services from both buyers as well as sellers. It is a place with various varieties are provided that can help the consumer to choose from and the buyer can get a platform where he can show case his product and get millions of the customer at the same time and he does not have to look for site all the time, it’s the system that take care of it. Now big data is playing a vital role in e commerce as it reads about user behavior and provides him a suitable product that he may need according to his behavior and query. There are various machine learning algorithms that are working on this and improving the services. [11] Basically in this paper we will read the user information and combine it with the product attributes and get a suitable suggestion for the user that will be most likely to be purchased by him. In the existing system we just look at one part of the case and give suggestion but in this paper we looked at both the sides, that is we looked after the product entities (the attributes and features that it poses) and the user behavior (the information given by the user and its previous history) that will better prediction and improve the system. Moreover for the optimized working of the system we included an enhanced version of HPCA scheduling algorithm for the Hadoop distributed file system also known as HDFS, which is very suitable for the heterogeneous system, the existing algorithm looks after the overall capacity of the node and then the tasks were assigned but here we will consider the health and the left over capacity of the nodes and arrange the queue for the same which will be refreshed all the time after the task is completed by any node.[18] The aim of the paper is to provide fast and most suitable suggestions to the users which can play a vital role in improving the sales of the company and getting the target done soon and faster

International Journal of Recent Technology and Engineering 2

Optimized Recommendation System for E-Commerce on Product Features and User Behavior

Recommender systems are information filtering software which is capable of resolving the recent issue of internet’s information overload. The recommender system generate the recommendation more suitably based on the data gathered either implicitly like user profile, click information, web log history or explicitly like ratings (scale 1–5), likes, dislikes, feedbacks. The most important challenge to the recommender system is the growing number of online users making it a high dimensional data which leads to the data sparsity problem where the accuracy of recommendation depends on the availability of the data. In this paper, a new approach called formal concept analysis is employed to handle the high dimensional data and a FCA-based recommender algorithm, User-based concept clustering recommendation algorithm (UBCCRA) is proposed. The UBCCRA out performs by accurately generating the recommendation for the group-based users called cluster users. The experimental result is shown to prove the cluster recommendation with good result. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2019.

Department of Multimedia

School of Design

Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST

A fca-based concept clustering recommender system

Big data is an idea of informational collections that depicts huge measure of information and complex that conventional information preparing application program is lacking to manage them. Presently, big data is a widely known domain used in research, academic and industries. It is utilized to store substantial measure of information in a solitary brought together one. Challenges integrate capture, allocation, analysis, information precise, visualization, distribution, interchange, delegation, inquiring, updating and information protection. In this digital world, to put away the information and recovering the data is enormous errand for the huge organizations and some time information ought to be misfortune due to circulated information putting away. For this issue the organization individuals are chosen to actualize the huge information to put away every one of the information identified with the organization they are put away in one enormous database that is known as large information. Remote sensor is a science getting data used to distinguish the items or break down the range from a separation. It is anything but difficult to discover the question effortlessly with the sensor. It makes geographic data from satellite and sensor information so in this paper dissect what are the structures are utilized for remote sensor in huge information and how the engineering is vary from each other and how they are identify with our investigations. This paper depicts how the calamity happens and figuring consequence of informational collection. And applied a seismic informational collection to compute the tremor calamity in view of classification and clustering strategy. The classical data mining algorithms for classification used are k-nearest, naive bayes and decision table and clustering used are hierarchical, make density based and simple k-means using XLMINER and WEKA tool. This paper also helps to predicts the spatial dataset by applying the XLMINER AND WEKA tool and thus the big spatial data can be well suited to this paper.