N. L. Bhanu Murthy

The polytope of degree sequences of hypergraphs

Abstract Let D n ( r ) denote the convex hull of degree sequences of simple r -uniform hypergraphs on the vertex set {1,2,…, n }. The polytope D n (2) is a well-studied object. Its extreme points are the threshold sequences (i.e., degree sequences of threshold graphs) and its facets are given by the Erdos–Gallai inequalities. In this paper we study the polytopes D n ( r ) and obtain some partial information. Our approach also yields new, simple proofs of some basic results on D n (2). Our main results concern the extreme points and facets of D n ( r ). We characterize adjacency of extreme points of D n ( r ) and, in the case r =2, determine the distance between two given vertices in the graph of D n (2). We give a characterization of when a linear inequality determines a facet of D n ( r ) and use it to bound the sizes of the coefficients appearing in the facet defining inequalities; give a new short proof for the facets of D n (2); find an explicit family of Erdos–Gallai type facets of D n ( r ); and describe a simple lifting procedure that produces a facet of D n +1 ( r ) from one of D n ( r ).

Impact of Feature Selection Techniques on Bug Prediction Models

Several change metrics and source code metrics have been introduced and proved to be effective features in building bug prediction models. Researchers performed comparative studies of bug prediction models built using the individual metrics as well as combination of these metrics. In this paper, we investigate whether the prediction accuracy of bug prediction models is improved by applying feature selection techniques. We explore if there is one algorithm amongst ten popular feature selection algorithms that consistently fares better than others across sixteen bench marked open source projects. We also study whether the metrics in best feature subset are consistent across projects.

Mining GitHub for Novel Change Metrics to Predict Buggy Files in Software Systems

Code change metrics mined from source control repositories have proven to be the most reliable predictors of bugs in contemporary software engineering research. Yet a definitive modus operandi for obtaining the required data from a particular software configuration management (SCM) repository needs to be put forward. In this paper, we define a modus operandi to extract some popular change metrics from the Eclipse repository on Github, which can be generalized for any open source Github repository. We define few code change metrics that are intuitively significant for predicting bugs. Bug prediction models built with these metrics along with the existing prominent code change metrics prove to be competent and consistent as per our experiments on five different versions of Eclipse JDT project. We explored Naïve Bayes Tree algorithm to build a prediction model and have found it to perform better than other commonly used algorithms in this problem domain.

Impact of Bug Reporter's Reputation on Bug-Fix Times

Software Analytics is gaining momentum as a result of involved empirical research in enhancing quality and productivity of software engineering activities. There have been rigorous research efforts in the areas of bug prediction and testing effort prediction by making use of historical data. The problem of predicting bug fix times is an interesting problem with lots of advantages to industry but there have been limited efforts to solve this problem. We introduce a new feature, the score of bug reporter, to predict bug fix time. The prediction accuracies of proposed model built with the new feature along with the other prominent features are found to be better than existing models.

Comparative study on effectiveness of standard bug prediction approaches

Bug prediction research has been evolving quite rapidly but its applicability to IT industry is far from reality. The efficacy of object oriented metrics in bug prediction has been investigated in this work by evaluating the performance of bug prediction models built with various combinations of these metrics. The module-order models have been built using object oriented metrics and the limitations of standard measures like recall in evaluating the performance of bug prediction models have been exposed.

Effect of Feature Selection on Kinase Classification Models

Classification of kinases will provide comparison of related human kinases and insights into kinases functions and evolution. Several algorithms exist for classification and most of them failed to classify when the dimension of feature set large. Selecting the relevant features for classification is significant for variety of reasons like simplification of performance, computational efficiency, and feature interpretability. Generally, feature selection techniques are employed in such cases. However, there has been a limited study on feature selection techniques for classification of biological data. This work tries to determine the impact of feature selection algorithms on classification of kinases. We have used forward greedy feature selection algorithm along with random forest classification algorithm. The performance was evaluated by selecting the feature subset which maximizes Area Under the ROC Curve (AUC). The method identifies the feature subset from the datasets which contains the physiochemical properties of kinases like amino acid, dipeptide, and pseudo amino acid composition. An improvised performance of classification is noted for feature subset than with all the features. Thus, our method indicates that groups of kinases are classifiable with maximum AUC, if good subsets of features are used.

TelNEClus: Telugu Named Entity Clustering Using Semantic Similarity

Semantic similarity plays a significant role in many of natural language processing (NLP) and information retrieval (IR) applications. Most IR methodologies represent the documents using the vector space model (VSM) traditionally known as bag-of-words (BoW) hypothesis. The main disadvantage of BoW is that the grammatical and the structural information of words is not captured. In this paper, we have attempted to cluster named entities (NEs) extracted from Telugu corpus based on semantic similarity. We contend that for this sort of work, more suited VSM is distributional hypothesis which is usually applied for measuring word similarity using word-context matrix. In the word-context matrix, the row vector is words given in the corpus; here, it is a proper noun as most of NEs are proper noun, and column vector is context such as windows of words, grammatical information. The row vector in word-context matrix is constructed in two ways with two different feature sets: The first way is to represent each NE with unique row vector (Row Vector1) without considering different occurrences in a corpus, and the second way is to represent each NE with a set of row vectors (Row Vector2) considering different occurrences in a corpus. For Row Vector1 representation, classical similarity functions like cosine, scalar product, Jaccard can be utilized, but for Row Vector2 representation, we have generalized similarity functions to Sum-of-Sum and Sum-of-Max. Experimentally, Row Vector2 representation enhances the clustering results.

CapAct: A Wordnet-Based Summarizer for Real-World Events from Microblogs

Short messages from microblog streams often contain information about real-world events. Streams of related messages can be clustered and classified as events or non-events. Summarizing events from clusters of event related messages is a challenging task as the summary needs to be concise yet informational. We present a novel method of summarization of events from short messages. We also propose a method of creating a set of extensive reference summaries from manually created summaries for effective evaluation. We used standard ROUGE based metrics to compare the proposed summarizer with many existing baselines including a strong Hybrid-tfidf method. Our summarizer consistently outperformed others in F1-score with a margin of 11% in ROUGE-1 and 5% in ROUGE-2 over Hybrid-tfidf.

Empirical Study on the Distribution of Bugs in Software Systems

Many research studies in the past have shown that the distribution of bugs in software systems follows the Pareto principle. Some studies have also proposed the Pareto distribution (PD) to model bugs in software systems. However, several other probability distributions such as the Weibull, Bounded Generalized Pareto, Double Pareto (DP), Log Normal and Yule–Simon distributions have also been proposed and each of them has been evaluated for their fitness to model bugs in different studies. We investigate this problem further by making use of information theoretic (criterion-based) approaches to model selection by which several issues like overfitting, etc., that are prevalent in previous works, can be handled elegantly. By strengthening the model selection procedure and studying a large collection of fault data, the results are made more accurate and stable. We conduct experiments on fault data from 74 releases of various open source and proprietary software systems and observe that the DP distribution outp...

A Comparative Study of Named Entity Recognition for Telugu

In this paper, we apply three classification learning algorithms to Telugu Named Entity Recognition (NER) task and we present a comparative study between these three learning algorithms on Telugu dataset (NER for South and South-East Asian Languages (NERSSEAL) Competition). The empirical results show that Support Vector Machine achieves the best F-measure of 54.78% on the dataset.