KMKalyani Mali

Kalyani Mali

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kalyani Mali has a total of 7 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Concurrent Line Perpendicular Distance Functions for Contour Point Analysis

    by Ratnesh Kumar, Kalyani Mali

    The authors develop a novel concurrent line perpendicular distance function (CLPDF) base shape descriptor. The shortest distance from an object contour points to a family of multi-lines passes through a fixed point and provides insight into the functioning of object recognition, matching, registration, and analysis of two-dimensional binary shape silhouettes. With this method, the authors compute the perpendicular distance from each point of an object contour to the family of concurrent lines passing through a fixed point. The fixed point is the centre of gravity of a shape. Geometrically (translation and rotation) invariant object features for shape-preserving classification in pattern recognition and image processing are characterized by the perpendicular distance functions. To make CLPDF scale invariant, the authors introduce a normalization function for the CLPDF base shape descriptor. In the matching stage, they compute and analyse the Euclidean distance between the eigenvalues of two shapes corresponding to the CLPDFs. The CLPDF methodology in pattern recognition and classification gives an excellent discriminative power, which is demonstrated by the excellent retrieval performance that has been experimented on several shape benchmark datasets, including Kimia silhouettes, and MPEG7 CE-Shape-1 datasets.


  • Open Access

    Firefly-SVM predictive model for breast cancer subgroup classification with clinicopathological parameters

    by Suvobrata Sarkar, Kalyani Mali

    Background: Breast cancer is a highly predominant destructive disease among women characterised with varied tumour biology, molecular subgroups and diverse clinicopathological specifications. The potentiality of machine learning to transform complex medical data into meaningful knowledge has led to its application in breast cancer detection and prognostic evaluation. Objective: The emergence of data-driven diagnostic model for assisting clinicians in diagnostic decision making has gained an increasing curiosity in breast cancer identification and analysis. This motivated us to develop a breast cancer data-driven model for subtype classification more accurately. Method: In this article, we proposed a firefly-support vector machine (SVM) breast cancer predictive model that uses clinicopathological and demographic data gathered from various tertiary care cancer hospitals or oncological centres to distinguish between patients with triple-negative breast cancer (TNBC) and non-triple-negative breast cancer (non-TNBC). Results: The results of the firefly-support vector machine (firefly-SVM) predictive model were distinguished from the traditional grid search-support vector machine (Grid-SVM) model, particle swarm optimisation-support vector machine (PSO-SVM) and genetic algorithm-support vector machine (GA-SVM) hybrid models through hyperparameter tuning. The findings show that the recommended firefly-SVM classification model outperformed other existing models in terms of prediction accuracy (93.4%, 86.6%, 69.6%) for automated SVM parameter selection. The effectiveness of the prediction model was also evaluated using well-known metrics, such as the F1-score, mean square error, area under the ROC curve, logarithmic loss and precision–recall curve. Conclusion: Firefly-SVM predictive model may be treated as an alternate tool for breast cancer subgroup classification that would benefit the clinicians for managing the patient with proper treatment and diagnostic outcome.


  • Fragment of Binary Object Contour Points on the Basis of Energy Level for Shape Classification

    by Ratnesh Kumar, Kalyani Mali

    This paper addresses the problem of translation, rotation and scale invariant recognition and classification of objects. A new set of translation and rotation invariant features are introduced called sigma-pi bond model in a lattice. They are the magnitude of energy required to disintegrate the particle from an object also called bond dissociation energy. The scale invariance of the energy level features of an object is obtained by using systematic computation of Hu’s seven moments that are used in the matching stage of classification and analysis of binary object shape. The proposed algorithm has been tested on a variety of binary objects data sets, including Kimia silhouettes and MPEG-7 CE-Shape-1 databases. In all the experiments, our approach demonstrate effective performance compared with other algorithms.


  • Significant Itemset Mining with Support-Attractiveness Framework

    by Subrata Datta, Kalyani Mali

    Support based popular itemset mining often misguides the users in proper knowledge discovery in connection with many real world applications such as social network analysis, source identification of epidemic diseases etc. Support only expresses the frequency of the itemsets in the database. In reality, support-based pruning technique does not bother about the significance of the itemsets in the context of correlation or associability. Therefore, support-based pruning technique often ignores valuable itemsets. In addition, support-based pruning technique sometimes generates commonsense itemsets that has lack of unexpectedness. Furthermore, it promotes isolated items that have no utilization in association rule mining. To fill up the shortcomings of support-based pruning technique, this paper first introduces the concept of attractiveness of the itemsets and thereafter, develops an efficient support-attractiveness framework (SAF) for itemset pruning. An efficient algorithm called SAFApriori (Support-Attractiveness Framework based Apriori) is proposed for significant itemset mining based on the SAF. The proposed algorithm has been tested with real and synthetic datasets and the obtained results show the effectiveness of the proposed algorithm.


  • Artificial Bee Colony Inspired Density Based Clustering with efficient Parameter Selection

    by Ankita Bose, Kalyani Mali

    In this article, we have suggested an Artificial Bee Colony (ABC) inspired density based clustering. It can effectively deal with the issue of parameter selection as well as representation of varying densities of general density based clustering approaches. We plot the average density of the neighbours of each object in the data set. This unknown density distribution is then approximated using kernel density estimation. This helps us to identify the threshold, so that we can divide the density distribution of the data set into core and noise and we can also identify the cluster transitions. Including the parameter selection, the overall process is optimized using ABC. The superiority of the proposed approach has been delineated considering various data sets and by using different measurement techniques.