D. Venkataraman

Automated lung cancer detection by the analysis of glandular cells in sputum cytology images using scale space features

Out of all various types of lung cancers, adenocarcinoma is increasing at an alarming rate mainly due to the increased rate of smoking. This work aims at developing a sputum cytology image analysis system which identifies benign and malignant glandular cells. In our proposed system, we develop an automated lung cancer detection system which segments the cell nuclei and classifies the glandular cells from the given sputum cytology image using a novel scale space catastrophe histogram (SSCH) feature. Catastrophe points occur when pairwise annihilation of extrema and saddle happens in scale space. Unusual nuclear texture shows the presence of malignancy in cells, and SSCH-based texture feature extraction from nuclear region is done. From the input high-resolution image, the cellular regions are localized using maximization of determinant of Hessian, nuclei regions are segmented using K-means clustering, and SSCH features are extracted and classified using support vector machine and color thresholding. The experimental results show that the proposed method obtained an accuracy of 87.53xa0% which is better than Gabor filter-based gray-level co-occurrence features, local binary pattern, and complex Daubechies wavelet-based features. The results obtained are in accordance with the dataset classified by medical experts.

A hybrid approach for recommendation system with added feedback component

With the increasing E-Commerce and online shopping there is a need for recommendation systems which help the customers in decision making and to suggest potential goods of purchase. In domains such as automobiles there are many websites but most of them are not having enhanced recommendation systems to enable easy decision making. Thus we have taken the initiative of building a dataset with multiple parameters based on a survey of the communities needs using potential blogs and created a recommendation system using user based and item based collaborative filtering. In addition to the combined collaborative filtering techniques we propose a framework which includes a feedback analysis to improve the recommendation system. The enhanced model aids the customers in decision making. We have proposed the feedback system at two levels. One is external feedback where the comments are gathered from public platforms like social media and automobile websites. The other is internal feedback i.e. the feedback is taken from users who have been provided with recommended items. The opinions extracted from such varied comments broadens the system and results. Our proposed hybrid model with feedback analysis has improvised the current system by providing better suggestions to customers.

Stroke Detection in Brain Using CT Images

Computed tomographic (CT) images are widely used in the diagnosis of stroke. The objective is to find the stoke area from a CT brain image and also improve the visual quality. The proposed algorithm helps to detect the stoke part in the absence of radiologist or doctors. Seed region growing (SRG) technique is the most popular method for segmentation of medical images because of high-level knowledge of anatomical structures in seed selection process. The proposed method consists of three steps: preprocessing, feature extraction, and segmentation. Feature extraction is done based on texture using the Gabor filter, and segmentation is done using SRG algorithm.

Knowledge representation of university examination system ontology for semantic web

This paper is aimed to provide a brief description about the working and the processes involved with the system. And it is also focusing on the challenges during the deployment of ontology for university examination system. The development of semantic web also enhanced the creation and publishing of ontology and relationships in almost all domains. The ontology developed in this paper using protégé can be extended to model and create data and metadata elements which are helpful for developing university examination applications and for knowledge representation in semantic web.

Computer vision based feature extraction of leaves for identification of medicinal values of plants

Plants are considered as one of the greatest assets in the field of Indian Science of Medicine called Ayurveda. Some plants have its medicinal values apart from serving as the source of food. The innovation in the allopathic medicines has degraded the significance of these therapeutic plants. People failed to have their medications at their door step instead went behind the fastest cure unaware of its side effects. One among the reasons is the lack of knowledge about identifying medicinal plants among the normal ones. So, a Vision based approach is being employed to create an automated system which identifies the plants and provides its medicinal values thus helping even a common man to be aware of the medicinal plants around them. This paper discusses about the formation of the feature set which is the important step in recognizing any plant species.

Leaf Recognition Algorithm for Retrieving Medicinal Information

India has a vast history of using plants as a source of medicines. This science is termed as Ayurveda. But, sadly somewhere in the race of keeping up with medicinal science and technology, India as a country has lost its track in the field of Ayurveda. Researchers and medicinal practitioners today, in spite of knowing that allopathic medicines are made using certain plant extracts, are oblivious about the medicinal the properties of plants. This paper aims at eradicating this problem, and hence strives to help potential users make better use of plants with medicinal properties. The dataset consists of 300 images of different types of leaves. The classification of the leaves is done with the help of a decision tree. Our system is an easy to use application which is fast in execution too. The objective of doing this paper is to develop an application for leaf recognition for retrieving the medicinal properties of plants. The recognition of leaves is done by extracting the features of the leaves from the images. The primary stakeholders involved with this project are researchers, medical practitioners and people with a keen interest in botany. We believe that this application will be an important part of the mentioned stakeholders’ daily lives. The primary purpose that this paper serves is to solve the problem of not knowing the useful properties of many plants.

Engender Product Ranking and Recommendation Using Customer Feedback

In our day-to-day life we tend to buy products on the Internet. There are plenty of consumer reviews on the Internet. If a customer wants to know about a product, he sees the review and rating of the product given by the product users. In this case we come to know about the importance of rating and review of the product which impacts the product’s market value. This article proposes a framework for calculating an accurate rating using customer feedback. In particular, we first take the consumer review as an input then remove all common words by using the information retrieval concepts like stop word removal and stemming. The next step is parts of speech tagging and finding the opinion word extraction to the rest of the phrases. Then we have to match the keywords with the ontology and finally we develop a probabilistic aspect ranking algorithm to rank the product. We see elaborately about our concept in this article.

Forensic future of social media analysis using web ontology

Whenever some user posts on social media networks, knowingly or unknowingly their activity is registered in countless online repositories. This exposes parts of the data to be publicly available; as a consequence of this a forensic analyst can reveal past activities, reconstruct a biased timeline and recover deleted data of the suspect. Movie ontology maps the input from movies and videos of user via facebook graph API to produce adaptive results of user activity related to their entertainment activities. Movie ontology will demonstrate how to reconstruct the social behavior of the user from the file and how to interpret the gained information as a potential source of evidence in digital forensic investigations.

Resource tracking in institutions using semantic web

The semantic web is evolving and helping users to find, share, understand and combine the information and be processed by the automated tools. With the traditional web, it is a time-consuming and tedious task to find the necessary information. By using the semantic web, we obtain information in a way that is more efficient and fast and avoid all the unnecessary data. Semantic web helps to make intelligent decisions by understanding the meaning of data. Educational institutions have the requirement of keeping track of resources like students, faculty, visitors etc. In our work, we have recorded the information of various resources of the institution in Web Ontology Language (OWL) using Protege tool. Then the SPARQL Query is used to acquire useful information. This helps to keep track of the various required resources. The developed ontology can be extended and reused by other institutions based on the need.

Improvising knowledge based acquisition on Zoo Management system

The development of semantic web technologies assures a new incentive for researches over software engineering. Although the basic concepts of semantic web have a widespread tradition over engineering sector, it moreover becomes difficult for software engineers to look into the diversity of ontology-provided approaches. A realistic scenario Zoo Management is taken based on which an ontology is generated to provide more semantic information. We illustrate how these techniques can be put into practice using the modern Semantic Web development tool Protege, and indicate future possibilities.