Bhawna Nigam

Categorizing the Document Using Multi Class Classification in Data Mining

Classification is the process of dividing the data into number of groups which are either dependent or independent of each other and each group acts as a class. The task of Classification can be done by using several methods using different types of classifiers. But classification cannot be done easily when it is to be applied on text documents that is: document classification. The main purpose of this paper is to analyze the task multi-class document classification and to learn that how can we achieve high classification accuracy in the context of text documents. Naive Bayes approach is used to deal with the problem of document classification via a deceptively simplistic model: assume all features are independent of one another, and compute the class of a document based on maximal probability. The Naive Bayes approach is applied in Flat (linear) and hierarchical manner for improving the efficiency of classification model. It has been found that Hierarchical Classification technique is more effective then Flat classification. It also performs better in case of multi-label document classification. The dataset for the evaluation purpose is collected from UCI repository dataset in which some changes have been done from our side.

Analysis of Markov model on different web Prefetching and caching schemes

The World Wide Web is growing rapidly in terms of number of users and number of web application. With this growth the response time of retrieving the web document is increasing. Users experience on the internet can be improved by minimizing users web access latency. This can be done by predicting the next step taken by user towards the accessing of web page in advance, so that the predicted web page can be prefetched and cached. This prefetching and caching is useful for reducing departure of user from the website and improving the quality of service. In this paper three different schemes for web Prefetching and caching are proposed i.e. Prefetching only, Prefetching with Caching and Prefetching from Caching. Prediction of the next accessed web page for prefetching and caching is achieved by modeling the web log using Dynamic Nested Markov model. Dynamic Nested Markov model is analyzed on these three Prefetching and Caching schemes. Experiments have been conducted on real world data sets.

Generating a New Model for Predicting the Next Accessed Web Page in Web Usage Mining

World Wide Web is growing rapidly. So it is necessary to study the user web navigation behavior to improve the quality of web services, offered to the web user. Analysis of user web navigation behavior is achieved through modeling web navigation history. Markov model is widely used to model the user web navigation sessions. Lower-order Markov model provides high coverage, but with low accuracy. Higher-order Markov model give low coverage but high accuracy with more time complexity. In this paper a new way of structuring the Markov model is proposed named as Dynamic Nested Markov model for modeling the user web navigation sessions. Dynamic Nested Markov model uses the nesting concept, the higher-order Markov model is nested inside the lower-order Markov model. Through this nesting, the second-order Markov model is accommodated inside the first-order Markov model. In Dynamic Nested Markov model, all the advantages of lower-order model and higher-order model are achieved in one model. In this model focus is on time complexity and coverage of the prediction state. Result shows that the high coverage has achieved and time complexity has been reduced.

Network Intrusion Detection using Semi Supervised Support Vector Machine

The use of Internet is growing bit by bit and therefore huge amount of security threats faced in front of computer network system. Due to these threats secrecy of the information which is available in the network system is highly affected. To protect our network system from these threats, it becomes very important to build up a system that acts as a barrier between the network systems and the unessential security attacks. For the monitoring and detecting the intrusion (unwanted access), an Intrusion Detection Systems (IDS) were developed. But the expected performance and accuracy are not achieved by these systems. In this paper we propose a Semi Supervised Support Vector Machine (S3VM) to overcome these two concerns. The semi supervised SVM also overcomes the shortcoming of supervised SVM that require only labeled data for training the classifier. Semi Supervised Support Vector Machine is based on Self Training algorithm for semi supervised learning. The dataset used for training and testing purpose is NSL-KDD dataset. This model provides classification accuracy up to 90%.

Classifying the bugs using multi-class semi supervised support vector machine

It is always important in the Software Industry to know about what types of bugs are getting reported into the applications developed or maintained by them. Categorizing bugs based on their characteristics helps Software Development team to take appropriate actions in order to reduce similar defects that might get reported in future releases. Defects or Bugs can be classified into many classes, for which a training set is required, known as the Class Label Data Set. If Classification is performed manually then it will consume more time and efforts. Also, human resource having expert testing skills & domain knowledge will be required for labelling the data. Therefore Semi Supervised Techniques are been used to reduce the work of labelling dataset, which takes some labeled with unlabeled dataset to train the classifier. In this paper Self Training Algorithm is used for Semi Supervised Learning and Winner-Takes-All strategy is applied to perform Multi Class Classification. This model provides Classification accuracy up to 93%.

Generating all Navigational Test Cases using Cyclomatic Complexity from Design Documents for Mobile Application

This is new mobile world; lots for even the simplest mobile application there can be many Navigational Paths. The challenge at Design phase is to identify all possible paths, generate code and validations for each one of them. Generating all possible test paths and test cases are difficult to achieve and need lot of manual efforts. In this paper we tried to generate possible navigational path using Design documents, so that Development and QA team get all the possible number of test cases in Design Phase itself. This will help to Estimate and analyze the actual project scope and timeline. QA Team save time as they do not have to generate paths manually and thereby reducing Navigational bugs getting detected and ensuring enhanced Path Coverage. General Terms Test cases, Cyclomatic Complexity, Navigational Path, Mobile Application, Design Document.

Mining Association Rules from Web Logs by Incorporating Structural Knowledge of Website

of the basic problems with the Association Rule discovery is that when Mining Algorithms are applied on Web Access Logs, the total number of generated rules is found to be very large. For finding useful results from these rules, the analyzer needs to look into large rule-set. Moreover, the analysis of such rule-set also requires certain criteria for making decisions, i.e. a particular rule should be accepted or not. This ambiguity of acceptance or rejection of rules makes it very difficult to extract knowledge. Hence in order to get effective results with the minimized effort, number of rules should be less and valid. Therefore, the structural knowledge of Website is considered to solve the purpose, that plays an important role in pruning the invalid rules, thereby reducing the size of rule-set , and it is observed from the experiment that the number of rules have been successfully reduced.