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

School of Computer Engineering & Technology

In this paper, we attempt to suppress noise from mouse input due to involuntary tremors from patients with motor disabilities. In order to achieve this task, we used techniques like breakpoint detection and mean filtering to preprocess our data which was then fed into a deep recurrent neural network. Additionally, we tested various architectures and explored existing hardware and software based assistive technologies. © 2019, Springer Nature Switzerland AG.

Engineering and Technology

International Conference on Innovations in Bio-Inspired Computing and Applications

LSTM Approach to Cancel Noise from Mouse Input for Patients with Motor Disabilities

Usage of social sites like Facebook, Twitter is increasing rapidly. People are using these sites for getting feedback about any product or service. Social data is the best data which business analyst can use for getting analysis results. From social data the investor can predict in which products the users are more interested in or what changes the users want in service. The social data analysis will definitely increase the sale or profit gained by the investor. The use of machine learning algorithms for analysing market data will add more knowledge into the knowledge of investor. In this study, we have proposed a method for analysing market data collected from social sites. We have shown the behaviour of different machine learning algorithms against market data. It is found that AdaBoost algorithm with one level decision trees perform best against market data. We have used AdaBoost to boost the performance of decision trees and proved with results. Different phases of social media data mining are also explained in detail.

World University root node

Journal of Engineering and Applied Sciences

Boosting Decision Trees for Prediction of Market Trends

The self-organizational ability of ad-hoc Wireless Sensor Networks (WSN) have led them to be the most popular choice. Despite significant advancements in Wireless Sensor Networks, efficient energy consumption in the network remains one of the most important research challenges. Clustering sensor nodes and organizing them hierarchically have proven to be an effective method to provide better data aggregation and scalability for the sensor network while conserving limited energy. Topology control mechanisms promotes efficient energy consumption as it deals with the arrangement of the nodes. Topology acts as virtual backbone of the network, so a good topology control mechanism improves the overall network lifetime. There are various techniques for topology control such as Self-Organizing algorithms, Minimal Spanning Trees and Unit Disk Graphs which are taken as basis for the proposed algorithm. The proposed algorithm performs hierarchical clustering on the network in order establish optimal routing path from sensor nodes to sink node whilst conserving network resources.

2018 2nd International Conference on Electronic Systems, Signal Processing and Computing Technologies [ICESC-2018]

Optimal Routing Path Selection to Improve Data Transmission and Lifetime of Wireless Sensor Network

Application of artificially intelligent methods for predictions is a fairly old area, although it is also the one in which there is always room for improvement in performance and in consistency, given the escalating nature of information and the varying efficacy of prediction logics. A hybrid of simple statistical methods coupled with intelligent computing (here artificial neural networks) is most likely to yield the closest prediction values with modest error rates. We propose to build an analytical and predictive model for estimating the stock market indices. This model can guide any kind of a user with or without experience in the stock market to make profitable investments. The forecasting done is by way of three statistical algorithms and an adaptive, intelligent algorithm, thus making the process fairly robust. Training and testing the neural network will be done with two-month stock market index values for some of the companies listed with the Bombay Stock Exchange. A comparative result of the four algorithms is calculated, and the one with best precision is suggested to the user with a sale/buy/hold answer. © Springer Nature Singapore Pte Ltd. 2018.

Advances in Intelligent Systems and Computing

Prediction of stock market indices by artificial neural networks using forecasting algorithms

The NUTRIC score (Nutrition Risk in the Critically Ill) is a method designed to calculate the risk of critically ill patients developing adverse conditions that is likely be improved by optimal nutritional therapy. The objective of this study is to validate the NUTRIC score for aggressive nutrition therapy and identify critically ill patients that are most likely to benefit from macronutrients with an aim to improving mortality rate. A logistic model is proposed to identify the relationship between the observed parameters of NUTRIC score like age, APACHE II score, SOFA score and number of co-morbidities using statistical methods and further authenticate it. The relationship between these parameters is often aggravated in the ICU due to the hypermetabolic nature of critical illness. This care can be extended by use of IoT devices which may be wearable for monitoring patients health. It will help not only to keep track patient's health, but to improve how the physicians deliver care as well. New technology innovations can be leveraged in nutritional health care context to monitor patients remotely. © 2018 Association for Computing Machinery.

ACM International Conference Proceeding Series

Extending nutritional risk screening in critically ill patients using iot devices

WSEAS Transactions on Signal Processing

The compression of digital imaging and communications in medicine images using wavelet coefficients thresholding and arithmetic encoding technique

Economic Denial of Sustainability (EDoS) attack is one of the major web security attacks performed on cloud hosted websites that exploits cloud’s utility model by fraudulently consuming metered resources such as network bandwidth. In such attack, the malicious traffic imitates to be legitimate and hence goes undetected. A way to defend against such attack is to analyze the browsing behavior of the users and classify them. A training dataset to be used for this classification includes some features that are fuzzy which may lead to incorrect results. Hence, there is a need of feature selection mechanism that selects only important features from the feature set and discards the irrelevant one. This paper proposes to use fuzzy entropy based feature selection for classification of website users in EDoS defense. To evaluate the performance, the classification is done with and without doing feature selection. The classification accuracy shows that the proposed approach is capable of producing more accurate results with fewer features than original feature space. © Springer Nature Singapore Pte Ltd. 2018.

Communications in Computer and Information Science

Fuzzy entropy based feature selection for website user classification in EDoS defense

This paper illustrates a new method for depression detection using EEG recordings of a subject. It is meant to be used as a computerised aid by psychiatrists to provide objective and accurate diagnosis of a patient. First, data from the occipital and parietal regions of the brain is extracted and different channels are fused to form one wave. Then DFT, using FFT, is applied on the occipito-parietal wave to perform spectral analysis and the fundamental is selected from the spectrum. The fundamental is the wave with the maximum amplitude in the spectrum. Then classification of the subject is made based on the frequency of the fundamental using rule-based classifier. Detailed analysis of the output has been carried out. It has been noted that lower frequency of the fundamental tends to show hypoactivation of the lobes. Moreover, low-frequency characteristics have also been observed in depressed subjects. In this research, 37.5% of the subjects showed Major Depressive Disorder (MDD) and in all 80% of the subjects showed some form of depression. © 2018, Springer Nature Singapore Pte Ltd.

Cognitive depression detection methodology using EEG signal analysis

In the current scenario, there is no computing system in practice which understands or provides analytical insights for the language used in the psychological documents maintained by psychiatrists. Most importantly, there is no standardized language or methodology used in psychiatric practice to diagnose a psychological disorder. The main aim of our system is to provide a standardized platform for an assistance in diagnosis using Natural Language Processing techniques and Sentiment Inference. The system would suggest to the psychiatrist, the psychological disorder identified and its severity from the document text. The severity is obtained via a gradient score, which is computed by the semantic based gradient processing module - the core of our research. The psychiatrist can approve/correct the system diagnosis if necessary and reassign the gradient values for phrases contributing towards disorder severity. In an automated process, model gets re-trained after a particular threshold is reached, continually improving its performance. © 2018 IEEE.

2018 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2018

Intelligent AI Assisted Psychological Disorder Analysis Using Sentiment Inference

Nowadays in educational institutions teachers provide problem statement to students and students submit their solution in accordance with problem statement provided. Grading of these solutions done manually. It is a time-consuming task. To find out the correctness of the solution they required to run multiple test cases. Also, teachers have to ensure that each computer in the lab has editor and compiler to run the code. This paper will provide guidelines for system interface to teachers in which teachers are able to create new programming assignments and assign it to a particular class. Also, this system helps teachers for grading of programming assignments. It will provide a detailed report of class for each programming assignments. It will also reduce the overhead of installing the software required to edit and run the code. This the system provide an easy interface to students to edit and to compile programs. It will also provide the detailed analysis of evaluated assignments. © 2017 IEEE.