Md. Saiful Islam

I am a first-year PhD student at the University of Rochester Computer Science department. I am advised by Prof. Ehsan Hoque, and my works focus on using natural language processing and machine learning for social goods. Some of my current projects are on tele-monitoring of neuro-degenerative diseases (e.g., Parkinson's). I am also interested in representation learning, summarization, and topic modelling, especially when there are multiple modalities. You can find the list of publications in my google scholar profile.

Previously, I served as an Assistant Professor at the Department of CSE, BUET. I have completed my B.Sc. degree in 2017 and M.Sc. degree in 2020 from BUET. Apart from my academic activities, I love reading, travelling, and playing table tennis. You can visit my Goodreads profile. I am a big fan of Liverpool Football Club. Also, I am highly interested in philosophy and human behavior.


Research

Keyword Aware Influential Community Search in Large Attributed Graphs

Co-authors: Mohammed Eunus Ali, Yong-Bin Kang, Timos Sellis, Farhana Murtaza Choudhury
Tags: community search, social networks, keyword similarity, academic graph

We introduce a novel keyword-aware influential community query (KICQ) that finds the most influential communities from an attributed graph, where an influential community is defined as a closely connected group of vertices having some dominance over other groups of vertices with the expertise (described as a set of keywords) matching with the query terms (words or phrases). We first design KICQ that enables users to issue an influential CS query intuitively by using a set of query terms, and predicates (AND or OR). In this context, we propose a novel word-embedding based keyword similarity model that enables semantic community search, which substantially alleviates the limitations of exact keyword search based community. Next, we propose a new influence measure for a community that considers both the cohesiveness and influence of a community and eliminates the need for specifying values of internal parameters of a network. Finally, we propose two efficient algorithms for searching influential communities in a large attributed graph. We present detailed experiments and a case study to demonstrate the effectiveness and efficiency of the proposed approaches.

Status: Under review in Information Systems Journal. (Pre-print is available)

Continuous Monitoring and Smart Emergency Warning System for Cardiovascular Patients

Tags: health informatics, machine learning, smart devices

Mobile-based smart monitoring and emergency response system for cardiovascular patients of Bangladesh. Vital signs like heart rate, blood pressure recorded by smart wearable devices are processed and analyzed by machine learning models. The output is a severity rating of the patient's health status and an advance predictive warning.

Status: Ongoing research project funded by ICT Division, Bangladesh.

Automatic Detection of NoSQL Injection Using Supervised Learning

Co-authors: Md Rafid Ul Islam, Zakaria Ahmed, Anindya Iqbal, Rifat Shahriyar
Tags: NoSQL injection, database security, MongoDB, CouchDB, automatic detection, machine learning, supervised learning

With the advancement in big data, NoSQL databases are enjoying ever-growing popularity. The increasing use of this technology in large applications also brings security concerns to the fore. Historically, SQL injection has been one of the major security threats over the years. Recent studies reveal that NoSQL databases also have become vulnerable to injections. However, NoSQL security is yet to receive the attention it deserves from the industry or academia. In this work, we develop a tool for detecting NoSQL injections using supervised learning. To the best of our knowledge, our developed training dataset on NoSQL injection is the first of its kind. We manually design important features and apply various supervised learning algorithms. Our tool has achieved 0.93 F2-score as established by 10-fold crossvalidation. We also apply our tool to a NoSQL injection generating tool, NoSQLMap and find that our tool outperforms Sqreen, the only available NoSQL injection detection tool, by 36.25% in terms of detection rate. The proposed technique is also shown to be database-agnostic achieving similar performance with injection on MongoDB and CouchDB databases.

Status: Published in IEEE COMPSAC 2019 (Wisconsin, USA). [Read Article]

Teaching Experiences

January 2019

  • CSE102 Structured Programming Language Sessional
  • CSE110 Computer Programming Laboratory
  • CSE204 Data Structures and Algorithms Sessional
  • CSE406 Computer Security Sessional

July 2018

  • CSE313 Operating System
  • CSE314 Operating System Sessional

January 2018

  • CSE403 Digital System Design
  • CSE404 Digital System Design Sessional

July 2017

  • CSE321 Computer Networks
  • CSE404 Computer Networks Sessional

January 2017

  • CSE102 Structured Programming Language Sessional
  • CSE218 Numerical Methods
  • CSE314 Operating System Sessional
  • CSE462 Algorithm Engineering Sessional
  • CSE464 Computational Geometry Sessional

AI Training

Theoretical and hands-on expertise on traditional machine learning and deep learning with Python, offered by Hiperdyne Corporation, Japan.[Course Outline]
November 2018-March 2019

Awards & Honors

  • University Merit Scholarship (8 times)
  • Dean's List Scholarship (4 times)
  • Champion in "Crowdsourcing information for city roads improvement" category in BRACathon 2015.
  • External judge of the ICPC Asia Kolkata-Kanpur Onsite Regional Contest 2018.
  • Supervisor of Bangladesh site, Asia-Pacific Informatics Olympiad.
  • Chief Judge, Bangladesh Olympiad in Informatics 2018.

Major Projects

Result Processing for Online College Admission System of Bangladesh

Tools : JAVA, Oracle

Result processing of about 1.5 million applicants for admission into 7,447 colleges of Bangladesh. This was a government-funded project to automate the entire higher secondary admission system. I designed and implemented the core algorithms for result processing using JAVA and Oracle database.

Project code is confidential and a property of Bangladesh Education Board.

Bangladesh e-Government Enterprise Resource Planning

A government-funded project for the office automation of ICT Division and Planning Commission of the Peoples' Republic of Bangladesh. I am currently working as a technical consultant in this project.

Project code is confidential.

Spatial Event Organizer

Tools : PHP, Google Map API, Android

There are multiple customers and multiple workers in a system. A customer can order something online, then a worker needs to go to the task location from his current location, pick the order, then deliver this to the home address of the customer. Google map API was used to estimate the duration of such travels. The spatial organizer distributes tasks among the employees so that the throughput is nearly optimal and all the employees are fairly loaded. The project was an android application of which, I designed and implemented the back-end algorithms in PHP.

Project code is available here.

Fix My Street

Tools : PHP, Codeigniter Framework, MySQL

An android application developed for Dhaka City Corporation. The citizens can report road and transportation-related issues to authority using this app. This project was initially developed for BRACathon 2015, a software development contest. After being the champion, this project was funded by BRAC. I designed and implemented the back-end using the PHP Codeigniter framework.

Project code is available here.

Fantasy Cricket

Tools : PHP, MySQL

Fantasy cricket is a game where a user forms a virtual team consisting of the players from the real game and receives points based on the performance of those players in the actual game. I designed the database and implemented the back-end using the PHP Codeigniter framework.

Project code is available here.


Resume/CV

Click here to download.

Miscellaneous

I am an avid reader and I love to collect quotes. Here are some of my favorites.

Contact

Apt 505, University Park

New York 14620, United States.

Mail: mislam6 AT ur DOT rochester DOT edu