A Walk-through of Final Year Projects of Bachelor’s in Computer Engineering (2066 – 2069) batch – IoE, Pulchowk Campus

Institute of Engineering (IOE) is one of the top engineering institutions in Nepal with Pulchowk Campus being the central campus. Every student in the college develop a final year project which has become one of the key factors of judging the performance of students in the college. The students of 066 batch of Bachelors in Computer Engineering completed their respective final year projects recently. Coordination of all the projects was done by Deputy Head of Computer and Electronics Department, Dr. Aman Shakya.

Most of the projects done by 066 batch are found to be related to data mining with a few being that of Image Processing and a handful of other projects from other categories. Most of the projects are web-based with some projects also developed for mobile platform (Android and iPhone). We have made a list of all projects giving brief introduction to each of them along with tools and technologies used by them.
1) Census Data Analysis And Visualization
The project made analysis and visualization of census data of the year 2011 making use of big data technologies. This project focuses on three major objectives. The first objective is the analysis of districts and VDCs based on socio-economic status with clustering approach. The second objective is a comprehensive analysis of the factors influencing economically active population with association rule mining techniques. The final objective is to develop a GIS based statistical data dissemination tool to provide flexible data access, digital mapping and charting. The tool is currently implemented for the 2001 census data. The results found from this research provide interesting and useful information to government, environmentalists and relevant decision makers’ in the assessment of the relative performance of a particular geographic area.

Tools used : Hadoop, MapReduce, Java – Spring MVC, C/C++, PHP, Javascript, Groovy, Quantum GIS, SPSS
Team Members: Anish Pradhan, Rikesh Shrestha, Sulav Kafley, Udeep Shakya
Supervisor: Babu Ram Dawadi / Co-Supervisor: Wiseyak Solutions

2) Corporate Decision Support System for Telecommunications
A decision support system for telecommunications with the use of data warehousing and data mining technologies. CDSST is a Business Intelligence tool providing complete analytics of revenue, service usage, customer behavior along with prediction of various of these factors. The project had 4 main stages – ETL (Extraction, Transformation and Loading of Data), building of Data Warehouse, Reporting and finally Data mining. Korn shell and Python scripts was used in ETL process with Teradata as the database for warehousing. Similarly, Microstrategy was used for reporting. Customer segmentation was used for done using RFM (Recency, Frequency and Monetary) algorithm while CART (Classification and Regression Tree) was used for customer’s churn prediction.

Tools used : Teradata, MicroStrategy, Python, Redmine, Datawarehouse, Data Mining, R
Team Members: Aashish Acharya, Santosh Ghimire, Sudip Kafle, Supritam Raj Shrestha
Supervisor: Dr.Nanda Bikram Adhikari / Co-Supervisor: Yomari Inc. Pvt. Ltd.
More information including reports, presentation and source code

3) XBRL initiative to facilitate financial reporting to the office of Company Registrar (OCR) in Context of Nepal
The project was done as an internship for Office of Company Registrar and helped in automating the process of financial reporting. The aim of this research project was to build a prototype hat could help OCR and Nepal Government in adoption of XBRL in Nepal. XBRL is a data rich dialect of XML, which was build to facilitate Business Intelligence by allowing machine to machine communication of business data, and processing over business data. XBRL based financial reporting has been adopted in over 100 countries, so its time that we also make a move in Nepal.

Tools used: xml, python, xquery, php, jfreechart, existdb, altova, cakephp, AJAX, javascript, Excel macro, vb, xbrl
Team Members: Aayush Shrestha, Ashok Basnet, Sarvagya Pant, Umanga Bista
Supervisor: Dr. Aman Shakya / Co-supervisor: Office of Company Registrar
More information including reports, presentation and source code

4) Warehouse based intelligent banking transaction analysis system<
The project made use of data warehousing and data mining technologies for analysis and forecasting of data related to a bank. The end result was a business analysis tool that helped in making proper decision to the bank regarding revenue growth and customer satisfaction. The project had 4 main stages – ETL (Extraction, Transformation and Loading of Data), building of Data Warehouse, Reporting and finally Data mining. Bash shell scripts was used in ETL process with Oracle as the database for warehousing. Similarly, OBIEE (Oracle Business Intelligence Enterprise Edition) was used for reporting. Support Vector machine – a data mining algorithm was used for ATM card fraud detection while CART (Classification and Regression Tree) was used for customer’s churn prediction.

Tools used: Bash Shell, Oracle, OBIEE, R-studio, Oracle R Enterprise
Team Members: Bibek Subedi, Dinesh Subedi, Jivan Nepali, Laxmi Kadariya
Supervisor: Prof. Dr. Subarna Shakya / Co-supervisor: Yomari Inc. Pvt. Ltd.

5) Educational Data Mining in relation to educational statistics of Nepal
The aim of this project is to develop a tool to analyze the educational statistics published by the department of education. Calculation of the Educational Development Index (EDI) based on many indicators and rank the schools, districts so that it would be useful for the decision, policy makers to focus on the key areas that has not been on their eyes. Principal component analysis was used for calculating EDI. ID3 algorithm has been used to construct Decision Tree so as to classify schools based on the performance of the schools (Average marks on six major subjects ).Regression analysis has been carried out to get the related parameters in the analysis. Cluster analysis was also carried on the 7000 schools that participated on the SLC exams of 3 consecutive years:2066, 2067 and 2068. The project also provides API of different parameters( schools, teachers, enrollment, EDI, and so on ) in JSON format. The project is currently live at hamroschool.org.

Tools used: Python, Django, MySQL, github, R, Open Street Map, Ajax, Json, Javascript, d3.js, Data mining
Team Members: Roshan Bhandari, Sijan Bhandari, Subit Raj Pokharel, Sujit Maharjan
Supervisor: Bibha Sthapit / Co-supervisor: Young Innovations Pvt. Ltd.
More information including reports, presentation and source code

6) Kathmandu valley Intelligent traffic system
The team have developed a system with web and iPhone application for traffic jam analysis with techniques of image processing. The system analyzes real time traffic data from traffic control and CCTV video data. Based on traffic conditions, the system then suggests the best route(with alternates) between start and end location in a mobile device. A web interface has also been developed in this project for the purpose of real time data entry from Traffic control room. CCTV live video feeds are processed and analyzed to find real time traffic situations and congestion level at different locations. The project is hosted at kvits.net63.net

Tools used: Java, Image Processing, Objective C
Team Members: Bikash Maharjan, Bikram Thapa, Roshan Manandhar, Yugesh Shrestha
Supervisor: Dr. Arun Timalsina / Co-Supervisor: ebPearls

More information including reports, presentation and source code

7) Subscriber Data Mining for Business Reporting and Decision Making in Telecommunications
Yet another Business Intelligence related tool for Telecommunications. The project built a data mart and reporting and mining of data was used. Unlike other BI projects which used BI tools, it was used with conventional database – MySQL and reporting was done with javascript library – Hicharts. It was complete web based tool with Python with Django framework on the backend and R used for data analytics.

Tools used: Python, Django, R, Data mining, Javascript, hicharts
Team Members: Bishal Timilsina, Bishnu Bhattarai, Narayan Prasad Kandel, Niroj Karki
Supervisor: Baburam Dawadi / Co-Supervisor : Manoj Ghimire
More information including reports, presentation and source code

8) Work-flow analysis and decision support system
The primary characteristic of project is the automation of processes involving combinations of human and machine-based activities, particularly those involving interaction with IT applications and tools. This project has been developed to identify these functional areas and develop appropriate specifications for implementation in workflow products. It is intended that such specifications will enable interoperability between heterogeneous workflow products and improved integration of workflow applications with other IT services such as electronic mail and document management, thereby improving the opportunities for the effective use of workflow technology within the IT market, to the benefit of both vendors and users of such technology.

Tools used: ASP.NET – MVC, C#
Team Members: Dicchya Shakya, Dipika Agrawal, Sabina Shrestha, Sanjana Shrestha
Supervisor: Prof. Dr. Subarna Shakya / Co-supervisor: ebPearls
More information including reports, presentation and source code

9) Vehicle license plate detection and recognition
One of the few projects utilizing Image processing techniques for vehicle number plate recognition. Neural Network – an Artificial Intelligence based algorithm was mainly used for number plate recognition. The best thing about the project was that it was made to recognize number plates of different sizes that are common in vehicles in Nepal. Regarding the programming language used, the project used Java with ImageJ (IJ) library.

Tools used: Image Processing, Java, ImageJ (IJ) libary
Team Members: Love Shankar Shrestha, Promisha Mishra, Ravi Bhagat, Tanka Bdr. Pun
Supervisor: Dr. Sanjeev Prasad Pandey / Co-Supervisor : Manoj Ghimire

10) “Neuro-Fuzzy Based Investment decision tool for stock market and commodities
The neural-network based system helped in predicting stock-market and commodities with data obtained from Nepal. Though we could find many previous projects related to stock market prediction, this was probably the first one using data of Nepal’s Stock Exchange. This project’s prediction model comprises between Neuro-Fuzzy models and Regression Analysis. Several Neuro-Fuzzy models were experimented using different input/output parameters and association Rule Mining has been performed to generate industry wise relation rules. The project was completed with ASP.NET as the programming language.

Tools used: C#, ASP.NET, Neural Network
Team members: Roshan Giri, Sarit Adhikari, Sharad Shrestha, Suresh Lamichhane
Supervisor: Dr. Arun Timalsina / Co-Supervisor: Manoj Ghimire
More information including reports, presentation and source code

11) SMS based communication for eWIN system of UN WFP in disaster situations
The project was done as an internship for World Food Program(WFP) to have proper data collection from rural areas with the use of SMS services. It is difficult to collect data from rural services for organizations, NGOs and INGOs due to lack of proper Internet connection. An Windows Tablet app was developed so that the data can be sent to the server from tablet / mobile device via Internet if it is available or via SMS. The developed Windows Tablet Application is able to send SMS messages of any arbitrary length, which when received in the server gets converted to the necessary existing data format.

Tools used: Python, Django, Windows Tablet, SMS
Team members: Abhishesh Bikram Bhandari, Gautam Acharya, Sabin Bhandari
Supervisor: Dr. Aman Shakya / Co-Supervisor: UN World Food Program
More information including reports, presentation and source code

12) Name conflict resolution for company registration – a natural language processing approach
It has been a tidious manual task for office of company registrar to detect similar name and avoid them during company registration. The utilization of Natural Language Processing for Nepali names helped in automating the process. Talking about technology, naming conflict resolution is about comparing of words. There are many systems developed for this purpose and are used widely. The automation requires natural language processing, translation of languages, transliteration between languages. There are several constraints for the checking provided by the Office of Company Registrar (OCR). These constraints should be considered while comparing words. The words need to be tokenized, stemmed before they can be further processed.

Tools used: Programming Language: Java SE 7, Database: MySQL Server, Testing: JUnit testing, IDE: Eclipse Indigo
Team Members: Gaurav Goyal, Janardan Chaudhary, Nimesh Mishra, Sanat Maharjan
Supervisor: Prof. Dr. Shashidhar Ram Joshi / Co-Supervisor: Office of Company Registrar

More information including reports, presentation and source code

  • Bibek Subedi

    Nice effort and a fabulous job for other student and especially for juniors. Keep posting posts like it in future. Best of luck

  • Kailash Budhathoki

    Fantastic job guys (y). Oh btw you can check old projects at http://uioe.sourceforge.net/ too. Anyone ready to takeover and update uioe?

  • Kailash Budhathoki

    Fantastic job guys (y). Oh by the way, you can check projects from older batch at http://uioe.sourceforge.net/ too. If anyone would like to takeover and update uioe, just let me know.

    • http://technott.com sudip kafle

      Thanks Kailash Bro! uioe.sourforge.net was a great effort and a source of inspiration to launch flipkarma. In fact we have uploaded all the projects from uioe to http://flipkarma.com.

  • Saugat Singh

    came across this site looking for idea.. I think I got one, thanks

  • http://techxhub.com Shaurya Verma

    thank you guys

    seriously a lot of great stuff…