People‎ > ‎

Harshit Jain, MSc Student

"The question is not whether it will work. The question is, if it does work, how important will it be?" 
                                                                                                                                                                                                - Marc Andreessen

About Me 

I am from "Agra, India". I have completed my undergraduate studies (2010-2014) in computer science from Anand Engineering College, affiliated by Uttar Pradesh Technical University, India. I worked as a Technical Coordinator in CCED, India. Currently, I am pursuing my Masters in Computer Science from the University of Victoria. 

My research is currently focused on "Context-Aware Personalized Recipe Recommender" that includes Internet of Things, Cyber Physical System and Self Adaptive System. I have taken courses that are relevant for my research which includes:
    • Self-Adaptive and Self-Managing Systems
    • Data Mining
    • Distributed Computing
    • Overlay Peer to Peer Networking 
    • Operations Research
I strongly believe that a software engineer should be instinctive, innovative and 

Contact Info.

Engineering/Computer Science Dept. Building.
Rigi Lab - ECS 412
Victoria, BC. Canada
E-mail: harshit[at]
Rigi Lab Phone: +1 250 472-5865
motivational, have intellectual ability, quick to adapt and adopt new technologies, which is only possible with solid foundation and profundity in his subjects. I have an insatiable desire to learn new technologies and fields which have help me to stay updated. I have developed analytical and reasoning skills over the years by pursuing my strong interest in Computer Science. 

I have always wanted to explore the unexplored problems in day to day life and found them interesting to solve to make the life easier. My ideal vision of a career has always been a job that involves creativity and innovation. 

Apart from academics, my hobbies are: travelling, watching movies, listening music and cooking.

Current Research Project 
    Context-Aware Personalized Recipe Recommender (CAPRecipes)

In the past few years, the general work habits of people have changed dramatically, raising concerns about their well-being. Numerous health related problems have been observed from their health records such as obesity, diabetes, heart diseases, etc. It is mainly due to lack of awareness about what is healthy to eat, while considering their taste. Thus to resolve these problems, we proposed the context-aware personalized recipe recommender system, that exploits both general and dynamically changing personal contextual data in order to suggest better personalized recipes. With the assistance of this proposed system, it is hoped that user’s recipes browsing experience will be enhanced keeping in mind their taste preferences and health restrictions.

Academics Projects


Having the translation capability in any application gives it an edge over the others as it can be used by a large variety of audience. For our coursework (overlay peer-to-peer networking course), we focuses on creating a minimal Chatting tools that can prove our proof of concept and can give us enough insight about the issues behind translation. Polyglot in English stands for a person who can speak multiple languages and so does our application. We thus created a window based chatting application named "polyGlot" that is based on Client-server architecture and can support individual (between two people) and group chat (between multiple people). Our main focus was to implement translation functionality which is achieved by using Google translation API. Ten popular languages are chosen to support translation  which are:English, Chinese  simplified  &  traditional, French,  German,  Italian, Japanese, Korean, Russian and Spanish. Project Report is available here.

 "PolyGlot" Demo:                Individual Chat                  Group Chat

     Forex Market Analysis and Prediction (Using Data Mining)

In this project, we have analysed the Canadian exchange rate with respect to the US dollar. The currency exchange rate market concerned to buying, selling and exchanging currencies at current or determined prices. There are many factors which affect the currency exchange rate, but we are particularly using the main factors and they are : Inflation, GDP and Interest Rate. These factors are also get affected by the change in exchange rate values. In order to know, where to invest, we should need an accurate prediction for the future exchange rate values. To achieve this, we have applied data mining techniques on two types of dataset : Yearly dataset (includes above three main factors along with the exchange rate) and Monthly dataset (includes exchange rate only). We have used two algorithms for each dataset and they are: Auto-Regressive Moving Average with exogenous inputs (ARMAX) and Multilinear Regression by Excel for yearly dataset and Auto-Regressive Moving Average (ARMA) and Univarate Multilayer Perceptron using Weka for monthly dataset. For both of the datasets, we have found that ARMA outperforms the rest of the algorithms and we have made the prediction for the future using ARMA. Project Report is available here.

    Analysis and Evaluation of Translation Services using MapReduce

This project is the analysis and evaluation of translation services using the mapreduce framework. Translation services assist the users of different language origin to interact with each other without being bounded by the language constraints. Since Big data is a recent upcoming technology in the market, it becomes very difficult to perform effective analysis using the existing traditional techniques. That is why, we used the MapReduce, which is comparatively a lightweight framework, to translate the massive data. We want to compare and learn the issues such as latency, geographical location, and accuracy of the different translation APIs while translating the large size of data. Our proposed solution is described below which uses the AWS EC2 services to evaluate the translation services across geographically separated locations. From the evaluation results, after comparing two translation services with: Various file sizes Varying the number of Mappers Three different languages, and Two different locations (US-Oregon, Asia-Mumbai). We deduce that the Google translate API outperforms the Bing Translate API. Project Slides and GitHub link are available here.

Undergraduate Project  
     Automatic traffic control and surveillance system 

This project aims to have a check on vehicle riders who violates traffic rules by automating penalty based traffic system using RFID technology. In this project the system tracks the violators and notifies him the penalty by sending a text message. In this way the violator is liable to pay the requisite amount. The implementation of this project is done by using C# using .NET framework.