MIT-World Peace University (MIT-WPU) is a top educational institute located in Pune, India. It is recognized within&nbsp;UGC under Government of Maharashtra Act No. XXXV 2017.&nbsp;

It is part of the MIT group of institutions (established in 1983) which covers 10+ campuses across India and has more than 50,000+ students enrolled across various disciplines at a given point of time. The institute houses 15 schools ranging from Engineering &amp; Technology, Design, Pharmacy, Commerce, Governance among others. It offers undergraduate, postgraduate and PhD programs among different disciplines.&nbsp;

&nbsp;

MIT World Peace University

Web services are a simple but effective method of connecting to network resources. To offer end-users data from their virtualized resources, all modern organizations provide web service interfaces. With the increased usage of multi-channel communication modes (mobile phones, tablets, and laptops), web services are excellent for guaranteeing seamless data flow across all platforms. While Web services address many difficulties, there are a few issues that consumers or requesters have when they locate a web service that satisfies their requirements. With the increased use of online services, there is a greater demand for an effective and almost flawless web service recommendation system. Cold start, sparsity, and overspecialization are some of the problems with recommender systems. To address this issue, in this paper, collaboration filtering and demographic filtering based on models are proposed as a new approach. Using this technique, recommendations can be generated based on the demographic characteristics of users, such as age, gender, occupation, etc. We propose using web services based on the results of both tests. This paper attempts to propose a solution to the cold start problem by using a hybrid filtering method.

School of Computer Engineering & Technology

Engineering and Technology

Recent Advances in Material, Manufacturing, and Machine Learning

Web service recommendation based on user's demographic attributes

Autism Spectrum Disorder (ASD) is a neurological condition that can affect a person's language development, speech, cognition, and social skills for the rest of their lives. Prevalence of ASD is about 1 in 100 children in India under the age of 10, and nearly 1 in 8 has at least one neurodevelopmental condition according to the 2011 Census [24]. Machine/Deep Learning methods have proven helpful in order to quickly and accurately determine the risk of ASD. Early intervention in the child's preschool and primary school years can aid in the development of important social, communicative, functional, and behavioral skills. It has been observed that very few studies and research has been done on ASD in India, which is our motivation to take this as our project. The very sparse number of current studies show that there is no existing dataset for children in India. In this paper, we intend to first understand what autism spectrum disorder is, the symptoms it presents itself in, currently available diagnosis tests, treatment options and how AI is relevant in this domain. Further, the idea would be to understand the extent of ASD study and research done in India. We have pioneered ASD Research in India as we formulated a dataset of Indian children to kickstart ASD research in India. Out of the three approaches we tried, the CNN-RNN Video Classification approach where the CNN block has a pretrained feature extractor (Inception V3) to classify whether a child has ASD or not performed with 98.48 % train and 90.48 % test accuracy. We have further explained the wide future scope this research inquiry has to offer. Our research intends to help establish the need for a population-specific study in the detection and diagnosis of ASD in Indian children. © 2022 IEEE.

World University root node

2022 IEEE 3rd Global Conference for Advancement in Technology, GCAT 2022

AutiScan: Screening of Autism Spectrum Disorder Specific to Indian Region

Social media networks are becoming a new reliable source of information and news in recent days. Lots of opportunities are provided by social media. Social media provides all the necessary information required for event detection from text and images, identification of key posts and posters and devising a warning system in case of disasters. Event detection in social media is one of the important tasks in computer field. It facilitates in developing preventive measures for the companies. The event detection is gathering attention because social media is getting popular day by day. The approaches that have been proposed don’t represent the social network in a complete manner. Notable changes can be made in the social network’s context by defining an event as occurrence with enough momentum and force. Thus, our motive is to detect an event based on its necessary characteristics. This paper describes different types of event detection methods that have been proposed to find out hot events, key posts, and influential posters. Also, various models have been described for analysis of data coming from social media. © 2021, Springer Nature Singapore Pte Ltd.

Lecture Notes in Electrical Engineering

Event Detection from Social Media Using Machine Learning

Computer system is widely used across the globefor several purposes in different fields and failure in thissystem is natural. It is not acceptable as it might lead to lossof important data. There are several reasons which cancause the computer system or an application to crash or fail,thus causing loss of important data lead to redo all theactivity or task, user was performing before the systemcrashed. Our system proposes a solution to this problem bybacking all the data to the cloud and thus can be restoredwhen the system recovers from the crash, allowing toresume the task which the user was initially performing.

International Research Journal of Engineering and Technology (IRJET)

Application Backup And Restore Across Multiple Devices

Businesses are increasingly using web services to expose and exchange data over the web. Last few years have viewed tremendous growth in web service as a technology for inter application logic sharing and data retrieval. As web services have been able to provide simple solutions of very generic business problems it has got much higher acceptance and presence over the internet. Today internet is flooded with web services which give rise to problem of selection of web service that will best fit the business use case. To work through the problem of overflow of web services over internet a very efficient web service selection mechanism is need of the hour. In this paper we proposed a novel approach for web service recommendation system. © 2016 ACM.

ACM International Conference Proceeding Series

A novel approach for web service recommendation

Classification of imbalance data is the major challenge to the community these days. Machine learning algorithms can evolve a one-sided classifier when data is imbalance. The vital challenge in imbalance dataset problem is that sometimes the minority (tiny) classes are more useful, but standard classifiers tend to be biased toward the majority (huge) classes and ignore the tiny ones. In this paper we compared the existing methods to handle imbalance dataset and provide a new hybrid approach which will improve the accuracy of classifier on imbalanced data. © 2016 IEEE.

International Conference on Communication and Signal Processing, ICCSP 2016

A new approach for handling imbalanced dataset using ANN and genetic algorithm

Imbalanced data set, a problem often found in real world application, can cause seriously negativeeffect on classification performance of machine learning algorithms. There have been many attempts at dealingwith classification of unbalanced data sets. To handle the problem of imbalanced data is to re balance themartificially by oversampling and/or under-sampling.

International Journal of Computer Science and Mobile Computing

A survey on methods to handle imbalance dataset

With the popularization of digital cameras and therapid development of the Internet, number of images and videostaken and its broadcasting is growing explosively. The aestheticsevaluation of video media can be used as useful clue to improveuser satisfaction in many applications like search; broadcastingand recommendation. To automatically assess the quality ofvideos with a strong correlation with human perception is achallenging task. There are two different challenges that exist inthis task: First one is, there is no any publically benchmarkeddatasets available till date and second one is lack of standard videoshooting techniques.