S. K. Sahay

Aruna Govada, Shree Ranjani, Aditi Viswanathan et al.

With data sizes constantly expanding, and with classical machine learning algorithms that analyze such data requiring larger and larger amounts of computation time and storage space, the need to distribute computation and memory requirements among several computers has become apparent. Although substantial work has been done in developing distributed binary SVM algorithms and multi-class SVM algorithms individually, the field of multi-class distributed SVMs remains largely unexplored. This research proposes a novel algorithm that implements the Support Vector Machine over a multi-class dataset and is efficient in a distributed environment (here, Hadoop). The idea is to divide the dataset into half recursively and thus compute the optimal Support Vector Machine for this half during the training phase, much like a divide and conquer approach. While testing, this structure has been effectively exploited to significantly reduce the prediction time. Our algorithm has shown better computation time during the prediction phase than the traditional sequential SVM methods (One vs. One, One vs. Rest) and out-performs them as the size of the dataset grows. This approach also classifies the data with higher accuracy than the traditional multi-class algorithms.

Aruna Govada, Bhavul Gauri, S. K. Sahay

Support Vector Machines (SVMs) were primarily designed for 2-class classification. But they have been extended for N-class classification also based on the requirement of multiclasses in the practical applications. Although N-class classification using SVM has considerable research attention, getting minimum number of classifiers at the time of training and testing is still a continuing research. We propose a new algorithm CBTS-SVM (Centroid based Binary Tree Structured SVM) which addresses this issue. In this we build a binary tree of SVM models based on the similarity of the class labels by finding their distance from the corresponding centroids at the root level. The experimental results demonstrates the comparable accuracy for CBTS with OVO with reasonable gamma and cost values. On the other hand when CBTS is compared with OVA, it gives the better accuracy with reduced training time and testing time. Furthermore CBTS is also scalable as it is able to handle the large data sets.

Ashu Sharma, S. K. Sahay

Malwares are big threat to digital world and evolving with high complexity. It can penetrate networks, steal confidential information from computers, bring down servers and can cripple infrastructures etc. To combat the threat/attacks from the malwares, anti- malwares have been developed. The existing anti-malwares are mostly based on the assumption that the malware structure does not changes appreciably. But the recent advancement in second generation malwares can create variants and hence posed a challenge to anti-malwares developers. To combat the threat/attacks from the second generation malwares with low false alarm we present our survey on malwares and its detection techniques.

R. K. Roul, O. R. Devanand, S. K. Sahay

The dynamic web has increased exponentially over the past few years with more than thousands of documents related to a subject available to the user now. Most of the web documents are unstructured and not in an organized manner and hence user facing more difficult to find relevant documents. A more useful and efficient mechanism is combining clustering with ranking, where clustering can group the similar documents in one place and ranking can be applied to each cluster for viewing the top documents at the beginning.. Besides the particular clustering algorithm, the different term weighting functions applied to the selected features to represent web document is a main aspect in clustering task. Keeping this approach in mind, here we proposed a new mechanism called Tf-Idf based Apriori for clustering the web documents. We then rank the documents in each cluster using Tf-Idf and similarity factor of documents based on the user query. This approach will helps the user to get all his relevant documents in one place and can restrict his search to some top documents of his choice. For experimental purpose, we have taken the Classic3 and Classic4 datasets of Cornell University having more than 10,000 documents and use gensim toolkit to carry out our work. We have compared our approach with traditional apriori algorithm and found that our approach is giving better results for higher minimum support. Our ranking mechanism is also giving a good F-measure of 78%.

R. K. Roul, S. K. Sahay

Exponential growth of the web increased the importance of web document classification and data mining. To get the exact information, in the form of knowing what classes a web document belongs to, is expensive. Automatic classification of web document is of great use to search engines which provides this information at a low cost. In this paper, we propose an approach for classifying the web document using the frequent item word sets generated by the Frequent Pattern (FP) Growth which is an association analysis technique of data mining. These set of associated words act as feature set. The final classification obtained after Naïve Bayes classifier used on the feature set. For the experimental work, we use Gensim package, as it is simple and robust. Results show that our approach can be effectively classifying the web document.

R. K. Roul, S. K. Sahay

Search engine returns thousands of web pages for a single user query, in which most of them are not relevant. In this context, effective information retrieval from the expanding web is a challenging task, in particular, if the query is ambiguous. The major question arises here is that how to get the relevant pages for an ambiguous query. We propose an approach for the effective result of an ambiguous query by forming community vector based on association concept of data minning using vector space model and the freedictionary. We develop clusters by computing the similarity between community vectors and document vectors formed from the extracted web pages by the search engine. We use Gensim package to implement the algorithm because of its simplicity and robust nature. Analysis shows that our approach is an effective way to form clusters for an ambiguous query.