Multi domain sentiment classification approach using supervised learning
Digital info out there on the net is increasing day by day. As a result of this, the demand for tools that facilitate individuals to find and analyzing of these resources also are growing in variety. Text Classification, particularly, has been terribly helpful in managing the information. Text Classification is that the method of assignment language text to 1 or a lot of classes supported the content. Its several necessary applications within the globe. As an example, finding the sentiment of the reviews, denote by people on restaurants, movies and different such things area unit all applications of Text Classification. During this project, the focus has been ordered on Sentiment Analysis that identifies the opinions expressed in a very piece of text. It involves categorizing opinions in text into classes like 'positive' or 'negative'. Existing works in Sentiment Analysis centered on decisive the polarity (Positive or negative) of a sentence. This comes below binary classification, which suggests classifying the given set of components into 2 teams. The aim of this analysis is to handle a unique approach for Sentiment Analysis known as Multi category Sentiment Classification. During this approach the sentences area unit classified below multiple sentiment categories like positive, negative, neutral then on. Classifiers area unit engineered on the prognostic Model, that consists of multiple phases. Analyses of various sets of options on the info set, like stemmers, n-grams, tf-idf then on, are thought of for classification of the info. Totally different classification models like Bayesian Classifier, Random Forest and SGD classifier area unit taken into thought for classifying the info and their results area unit compared. Frameworks like woodhen, Apache driver and sickest area unit used for building the classifiers
Published by: Parineeta Jha, Sajid Khan
Author: Parineeta Jha
Paper ID: V5I2-1167
Paper Status: published
Published: March 6, 2019
Content based Image Retrieval System using K-Means Clustering Algorithm and SVM Classifier Technique
The dramatic rise in the sizes of pictures databases has blended the advancement of powerful and productive recovery frameworks. The improvement of these frameworks began with recovering pictures utilizing printed implications however later presented picture recovery dependent on substance. This came to be known as Content Based Image Retrieval or CBIR. Frameworks utilizing CBIR recover pictures dependent on visual highlights, for example, surface, shading and shape, rather than relying upon picture depictions or printed ordering. In the proposed work we will use various types of image features like colour, texture, shape, energy, amplitude and cluster distance to classify the images according to the query image. We will use multi-SVM technique along with clustering technique to compare the features of the input image with the input dataset of images to extract the similar images as that of the query image.
Published by: Harleen Kaur Maur, Puneet Jain
Author: Harleen Kaur Maur
Paper ID: V5I2-1143
Paper Status: published
Published: March 6, 2019
Comparison of recovery profile of sevoflurane and propofol as induction agent in day care surgery
Daycare surgical procedure is those procedures which are performed in a hospital or an outpatient setting/surgeon’s office where the patient is discharged within 24 hours. 1% Propofol and 8% sevoflurane are commonly used induction agents for such procedures. It is of great importance to select a better induction agent with rapid onset and recovery with minimal side effects. Our study comprises of 60 cases undergoing various daycare surgical procedures that were admitted in National Institute of Medical Sciences and Research Hospital, Jaipur from January 2017-June 2018. This study was carried out to compare which out of propofol and sevoflurane is a better induction agent for day care procedures. In our study, we found that induction was faster and with fewer complications with propofol as compared to sevoflurane which was associated with a higher incidence of postoperative nausea and vomiting. We concluded that propofol is a better induction agent than sevoflurane for daycare surgical procedures.
Published by: Robin Lohia, Dr. Meenaxi Sharma
Author: Robin Lohia
Paper ID: V5I2-1165
Paper Status: published
Published: March 5, 2019
Forward secure ID based ring signature for data sharing
Cloud computing provides services where one can access information from any place, from anywhere, at any time. So basically cloud computing is subscription based service where one can obtain network storage space and computer resources for data storage as well as data sharing. Due to high fame of cloud for data storage and sharing, a large number of participants gets attracted to it. The security is the biggest concern for the adoption of the cloud. The major issues in this regard are efficiency, data integrity, privacy, and authentication. In order to handle these issues concept of a ring, the signature has been introduced for data sharing amongst a large number of users. Ring signatures are used to provide user’s anonymity and signer’s privacy. But the expensive certificate verification within the ancient Public Key Infrastructure (PKI) setting becomes a bottleneck for this solution to be scalable. ID-based ring signature had been introduced which eliminates the process of certificate verification. Further enhancement of security with forwarding security concept has been introduced. According to this idea, if a secret key of any user has been compromised; all previously generated signatures that embrace this user still stay valid. This property is very vital to any giant scale knowledge sharing system because it is not possible to raise all knowledge data owners to re-authenticate their data whether or not a secret key of 1 single user has been compromised. Thus we propose a secure ID-based ring signature with forwarding security.
Published by: Swati Khatal, Tabassum Maktum
Author: Swati Khatal
Paper ID: V5I2-1153
Paper Status: published
Published: March 5, 2019
Recommendation of food tourism using Artificial Neural Network – A survey
According to “Global Report on Food Tourism” of the World Tourism Organization (UNWTO), food tourism is commented as a fast-growing segment of the tourism industry. Food tourism means traveling to seek enjoyment via eating and drinking experiences at the destination. movies. Such a system can suggest a set of recipes to users based on their interest, or the popularities of the dishes. For tourism services like"TripAdvisor" , the most successful online forum in the travel and tourism industry which recommends the food items based on the search history. Here, we using the ANN algorithm to predict that which algorithm is better to analyses the datasets of food recipes. Finally, it can make some recommendations for the tourist with better-personalized traveling experience and food services.
Published by: Suruthi P., Varsa Sukirthana S., Raguvaran S.
Author: Suruthi P.
Paper ID: V5I2-1152
Paper Status: published
Published: March 5, 2019
Advanced billing system for government departments
Making payments for electricity, water, a property is more of a routine job which needs to be done but often delayed mainly due to its tedious nature. Also getting access to various details of the same is also tedious. Earlier the existing system was manual payment methods where the user needed to visit the nearby payment center and pay their respective bills. The proposed system is more secure, error-free and easily incorporable to any further developments and changes to building an application program to reduce the manual work of managing amount of units consumed by the costumers and generating the bills for various departments like the municipality, electricity and water department.
Published by: Ranjitha H. T., Chethana M.
Author: Ranjitha H. T.
Paper ID: V5I2-1161
Paper Status: published
Published: March 5, 2019
