@nitgoa.ac.in
NIT Goa
Ph.D. from NIT Goa, India.
Data Mining and Analysis, Social Media Mining
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Seema Wazarkar, Bettahally N. Keshavamurthy, and Evander Darius Sequeira
Springer Nature Switzerland
Ajit Dhanawade, Seema Wazarkar, and Shailendra Kumar
Springer Science and Business Media LLC
Seema Wazarkar, Ketan Kotecha, Shruti Patil, and Nidhi Kalra
IEEE
Social image content analysis is one of the important tasks for fashion analysis. Use of proposed system in fashion industries will uplift their business as social visual perception is very useful for the decision making in fashion industries. It supports growth in business and helps in minimizing loss or provides prior knowledge of risks. Analysis of social content data is a challenging task due to the nature of social data. Social content data is unstructured and full of ambiguity. But, this source of data is very important because it keeps updating continuously so that current data is being available for analysis. There is a necessity of current data for applications related to fashion as fashion trends keep changing. Therefore, in this paper convolutional neural networks is applied along with our machine learning approaches to find optimal fashion analyzing approach where social media is utilized to predict fashion style. Deep learning approach with Softsign and Softplus function performed well.
Vaibhav Kadam, Satish Kumar, Arunkumar Bongale, Seema Wazarkar, Pooja Kamat, and Shruti Patil
MDPI AG
In the era of Industry 4.0, the idea of 3D printed products has gained momentum and is also proving to be beneficial in terms of financial and time efforts. These products are physically built layer-by-layer based on the digital Computer Aided Design (CAD) inputs. Nonetheless, 3D printed products are still subjected to defects due to variation in properties and structure, which leads to deterioration in the quality of printed products. Detection of these errors at each layer level of the product is of prime importance. This paper provides the methodology for layer-wise anomaly detection using an ensemble of machine learning algorithms and pre-trained models. The proposed combination is trained offline and implemented online for fault detection. The current work provides an experimental comparative study of different pre-trained models with machine learning algorithms for monitoring and fault detection in Fused Deposition Modelling (FDM). The results showed that the combination of the Alexnet and SVM algorithm has given the maximum accuracy. The proposed fault detection approach has low experimental and computing costs, which can easily be implemented for real-time fault detection.
Seema Wazarkar and Bettahally N. Keshavamurthy
Elsevier BV
Abstract Fashion industries need to be attentive towards the changing fashion and its upcoming market demands to grow their business, optimally. This paper describes research work involved in image mining for fashion analysis and forecasting using fashion-related images collected from the social network. A novel soft clustering technique is proposed for grouping the social fashion images. This technique is robust against uncertainty found in given images. The proposed clustering approach is compared with existing soft clustering approaches. It is found that the proposed approach performs well. Attributes of fashion items found in each cluster are analyzed through correlation, causal analysis, and fashion cycle visualization. Predictive models are applied to the clustered fashion items for style forecasting. A comparative study of predictive models is also done to find an optimal technique for various fashion items. As social visual perception is helpful for decision making, the proposed system is very useful in fashion industries to uplift their business.
Seema Wazarkar and Bettahally N. Keshavamurthy
Springer Science and Business Media LLC
Social image mining is beneficial to accomplish tasks like event detection, suspicious activity detection, prediction of future trends, identification of mentally depressed people, etc. To carry out social image mining, data mining techniques need to be used. Clustering is one of the most important tasks of data mining which is able to deal with the unlabelled data. But, less number of clustering approaches are having ability to handle the uncertain image data. Thus, in this paper we proposed a soft clustering algorithm named as ROugh Mean Shift clustering (ROMS) with layered feature extraction model for social images. Effectiveness of the rough set theory and mean shift concepts are incorporated in this algorithm. It makes the ROMS to deal with the vagueness and the automatic determination of cluster numbers in given data. Proposed method is experimented on three datasets- synthetic, standard and real-world datasets and compared with existing techniques. Experimental results show that ROMS performs better as compared to other techniques.
Ahsan Hussain, Bettahally N. Keshavamurthy, and Seema Wazarkar
IGI Global
Information-disclosure by social-users has increased enormously. Using this information for accurate location-prediction is challenging. Thus, a novel Multi-Layer Ensemble Classification scheme is proposed. It works on un-weighted/weighted majority voting, using novel weight-assignment function. Base learners are selected based on their individual performances for training the model. Main motive is to develop an efficient approach for check-ins-based location-classification of social-users. The proposed model is implemented on Foursquare datasets where a classification accuracy of 94% is achieved, which is higher than other state-of-the-art techniques. Apart from tracking locations of social-users, proposed framework can be useful for detecting malicious users present in various expert and intelligent-system.
Ahsan Hussain, Bettahally N. Keshavamurthy, and Seema Wazarkar
Springer Singapore
The rise of social media and social platforms has led to enormous information dissemination. Images shared by social users at any moment convey what they see or where they have been to. Social images express far more information than texts which may involve individuals’ characteristics like personality traits. Existing methods perform event classification based on fixed temporal and spatial image resolutions. In this paper, we thoroughly analyse social network images for event classification using Convolution Neural Networks (CNNs). CNN captures both important patterns and their contents, to extract the semantic information from the images. We collect shared images from Flicker specifying various sports events that users attend. Images are divided into three event classes, i.e. bikes, water and ground. After extensive experimentation using CNN, for training and classifying images, we obtain an accuracy of 98.7%.
Seema Wazarkar and Bettahally N. Keshavamurthy
Springer Science and Business Media LLC
Social image data related to fashion is flowing through the social networks in huge amount. Analysis of this data is a challenging task due to its characteristics like voluminous, unstructured, etc. Classification provides an easy and efficient way to deal with such data. In this paper, we proposed a new approach for classification of fashion images by incorporating the concepts of linear convolution and matching points using local features. Linear convolution is used to get the representative images with important features. Then, matching points between given image and class representative images are obtained. Maximum matching points are considered while assigning a class label to the given image. Proposed approach is useful further for various applications related to fashion such as fashion recommendation, fashion trend analysis, etc.
Bettahally N. Keshavamurthy, Shashank Prakash Srivastava, Jaseel Haris, Ankush Kumar, and Seema Wazarkar
IEEE
Word2vec is an assortment of related models specially employed to yield word embeddings. By its application to a relatively large dataset that corresponds to a given event coming about at a given point of time at a given location, we can break down the event into sub-events, and study them further. Investigating sub-events in the right direction can help us in countless ways. It can enable us to decipher their local yet inevitable impacts which might otherwise have gone missing in the sea of the whole event altogether. In our paper, we have broken down the event (of the happenings of 'Kashmir') into sub-events and pulled out a few randomly. We have then applied sentiment-analysis to each one of them instead of applying it on to the whole event all at once. The rise and fall of the sentiment with respect to each sub-event is plotted and the variation is visualised in the end. The procedure is not just limited to our domain of interest but can be adopted to study any event.
Seema Wazarkar and Bettahally N. Keshavamurthy
Elsevier BV
Abstract A huge amount of image data is being collected in real world sectors. Image data analytics provides information about important facts and issues of a particular domain. But, it is challenging to handle voluminous, unstructured and unlabeled image collection. Clustering provides groups of homogeneous unlabeled data. Therefore, it is used quite often to access the interesting data easily and quickly. Image clustering is a process of partitioning image data into clusters on the basis of similarities. Whereas, features extracted from images are used for the computation of similarities among them. In this paper, significant feature extraction approaches and clustering methods applied on the image data from nine important applicative areas are reviewed. Medical, 3D imaging, oceanography, industrial automation, remote sensing, mobile phones, security and traffic control are considered applicative areas. Characteristics of images, suitable clustering approaches for each domain, challenges and future research directions for image clustering are discussed.
Seema Wazarkar, Bettahally N. Keshavamurthy, and Ahsan Hussain
IGI Global
In this article, probabilistic classification model is designed for the fashion-related images collected from social networks. The proposed model is divided into two parts. The first is feature extraction where six important features are taken into consideration to deal with heterogeneous nature of the given images. The second classification is done with the help of probability computations to get collection of homogeneous images. Here, class-conditional probability of extracted features are calculated, then joint probability is used for the classification. Class label with maximum joint probability is assigned to the given image. A comparative study of proposed classification model with existing popular supervised as well as unsupervised classification approaches is done on the basis of obtained accuracy of the results. The effect of convolutional neural network inclusion in the proposed feature extraction model is also shown where it improves the accuracy of final results. The output of this system is useful further for fashion trend analysis.
Seema Wazarkar, Bettahally N. Keshavamurthy, and Ahsan Hussain
Elsevier BV
Abstract Social image data is very useful to solve many real world problems. In this paper, a novel soft classification approach is proposed to deal with the problem of vision segmentation in social networks. Proposed approach is inspired by the k-nearest neighbour and soft classification concepts. K-nearest neighbour is one of the popular and simplest classification algorithms. Soft classification has provision for assigning more than one class label to a single object. Here, soft classification concept is incorporated in k-nearest neighbour algorithm to detect the ambiguous regions of the image. Experimentation is carried out on the images collected from social networks. Three social image datasets i.e. synthetic, standard and real-world are used. Proposed approach performed much better as compared to the traditional k-nearest neighbour approach. It is useful for accomplishing various tasks like fashion analysis, emotion detection, event detection, etc. through object detection and recognition.
Seema Wazarkar and Bettahally N. Keshavamurthy
Springer Singapore
Extraction of appropriate features is a difficult task because it mainly depends on a specific application domain. In this paper, we presented a 5-layered feature extraction model for social images. This model extracts color, texture, geometric, and regional features from given image and also checks presence or absence of people in an image by face detection. Then, normalization of the feature vector is done with the help of priority element. Proposed model is able to deal with the heterogeneous nature of social images. It is useful to get good results in the field of social data analytics.
Seema V. Wazarkar and Amrita A. Manjrekar
IEEE
Text clustering is advantageous for extraction of text data from web applications such as e-news papers, collection of research papers, blogs, news feeds at social networks, etc. This paper presents a text clustering Hierarchical Fuzzy Relational Eigenvector Centrality-based Clustering Algorithm (HFRECCA). The algorithm is a combination of fuzzy clustering, divisive hierarchical clustering and page rank algorithm. Travel guide articles are pre-processed to remove stop words and stemming. Then, similarity matrix is generated using word distance computation. In HFRECCA, divisive hierarchical clustering algorithm is applied where it uses Fuzzy Relational Eigenvector Centrality-based Clustering Algorithm (FRECCA) as sub routine algorithm. FRECCA outputs cluster membership values on the basis of page rank score using page rank algorithm and generate clusters according to it. HFRECCA has features of hierarchical clustering as well as fuzzy clustering as it creates hierarchy of clusters and an object can belong to multiple clusters. Structure of information resides in text documents is hierarchical hence HFRECCA is useful for clustering of data from natural language documents.