Grp No. Project members Title of Project Abstract Project Guide
  • Arnab Sanyal
  • Omkar Kadam
  • Rishabh Shetty
  • Marvell Periera
Investment Analysis using Event Detection on Alternative Data The proposed systems aim is to find the correlation between the real-time events and financial trends of a company. This data can then provide insights as to the overall performance of the company and how it has been affected by events in the past and thus the kind of effect a current event might have. The project is implemented over 3 stages : Data Extraction which involves information extraction,knowledge representation and reasoning using methods like bag-of-words and TFIDFs(Term Frequency Inverse Document Frequency) to store semantic data, Event Detection which involves training of a topic modeller to extract events ,name entity recognition to increase the accuracy of the event detection and Visualization which includes a time series plot of parameters of company and the correlation between the trend and event. Prof. Anagha Shastri
  • Ashish Sharma
  • Raina Lopes
  • Furqan Khan
  • Joel Keerickal
School GIS With the growing digital world, there is huge amount of data generated daily. Amongst which there is geospatial data available which can be useful for getting fruitful knowledge. Such information can be used for the betterment and providing insights about a system as well as highlight important areas which need attention. The purpose of this project is to create a community driven and oriented GIS portal which will help to contribute as well as gain information about schools in Maharashtra.
Currently there are two websites, first is school which maps schools and school report provides data about a particular school.The goal of our project is to not only provide school data but also provide correlations between different demographic aspects and performance of the schools, provide a platform for the community to give reviews, rate a school, provide and validate school data through crowd sourcing.
Prof. Vaishali Kavthekar
  • Vinayak Gaonkar
  • Nicholas D’souza
  • Pratik Rane
  • Catherin Johnson
Development of Community GIS system Mumbai is currently one of the fastest developing cities in the world. This development has been possible due to a well thought and executed set of plans. The latest urban development plan was released by the BMC in 2015. This data was made available to the public. The format, however, was not really in a manner that could be easily understood; or used. It was just raw data. Using GIS, this system aims at improving the quality of the provided data. The information will be digitalized and added to the survey maps provided by the BMC. This mapping will be done over in several steps and will require various software such as QGIS, GeoNode, Cartoview, ODK, etc. This platform will provide a much better and understandable data format. In such a system, incorrect data can be checked as the participation of the people leads to a better insight. The data provided was based on research from a foreign company and was not in complete sync with the actual state. All these issues can be resolved with this system and it will also help keep track of the development taking placed and to check what tasks were promised, have been fulfilled, are being fulfilled or are being looked over. This technology could revolutionize the implementation of the urban Prof. Tayyabali Sayyad


Grp No. Project members Title of Project Abstract Project Guide
  • D'cruz Nicky
  • Chirayath Leo
  • Dias Steve
Diet Expert System Diet NLSS is an application which analyses user profile to recommend appropriate diet to the user The application uses users weight, height and age to calculate the appropriate calories required by the user for a day . Using this value the application generates a diet or the user is given an option to create his own diet from a pre - existing database. Optionally the user can also feed the information about his eating habits which can be used to generate a more personalized diet plan . User can also calculate his calories burnt through physical activity and integrates the results with the diet. User will also be provided with feedback through a graph which will help user to keep track of his activities throughout the week.The user is also notified through timely notifications about the timings of his Meal. Prof. Sushree S
  • Britto Franklin
  • Dias Godfrey
  • Fernandes Llewellyn
  • Pereira Calida
M2M for utility services With the increasing consumption of electricty and thus fossil fuels , it is the need of the hour that measures must be taken to reduce the same. Most measures to reduce consumption are restricted to being curative. This project deals with developing an energy monitoring system that will enable consumers to understand how and what they are being billed for. In addition o this , it will also forecast the consumers future bill based on their previous and present consumption rates. This will enable the customers to take measures to reduce their consumption before the end of the month ,thus saving electricity while lowering the consumers electricty bills. To summarize, it involves the development of a M2M application that monitors and reports the electricity usage of the consumers , enabling them to curtail their future usage based on this data. Prof. Mahalaxmi S
  • Bhalerao Deven
  • Jain Manish
  • Jha Vishal
  • Naik Kunal
Detection of Plagiarism in Software Source Code This tool is used for detecting the percentage of plagiarism of source code between two or more java files (i.e. .java). While detecting the plagiarism between the files it will use the three approaches i.e. Software metrics, Cosine Similarity and Machine Learning. In Software metrices approach it will check the structure of the files i.e. number of open and closed braces, function name, percentage of pure comment lines, average indentation in tabs and white spaces after open braces, program line length in term of characters, etc. In Cosine Similarity approach it will check the similarity between the files on the basis of frequency of unique word count. In Machine Learning approach it will generate rules and patterns on the basis of clusters formed. Finally it will generate the appropriate result from these approaches and on the basis of that result it will determined the percentage of plagiarism between the files. Prof. Prasad Padalkar