KDDM / Research areas

Four clusters,
one discipline.

Knowledge discovery and data mining is an interdisciplinary area focused on methods for extracting useful knowledge from data. The rapid growth of online data creates a pressing need for it, drawing on statistics, databases, pattern recognition, machine learning, data visualization, optimization and high-performance computing.

A

Social Media Data Mining

Social media data mining uncovers hidden patterns and trends in platforms such as Twitter, LinkedIn and Facebook, using machine learning, mathematics and statistics. Our work covers spam reviews and spammer groups, rumour and fake-news detection, app-store analytics and malware detection.

Selected publications

Theses

  • PhD FA21Naveed Hussain: Spam Review Detection through Behavioral and Linguistic Approaches
  • PhD FA25Ahmed Bilal: Investigating the Significant Features Influencing Success of Smart Phone Applications
  • PhDAsia Maqsood (in progress): Malware Detection and Classification using Machine Learning and Deep Learning Techniques
  • PhDWaqas Ali (in progress): A Study on Fake News Detection and its Propagation
  • MS SP22Aroma Javed: A Deep Learning based Prediction of Stock Market Trend using Social Media
  • MS SP22Zohaib Ahmad: Mobile App Fraud and Malware Detection using Data Mining based Methods
  • MS SP22Usman Bhatti: Repeated Buyer Prediction: A Study of Repurchasing Intention of Buyer in E-Commerce
  • MS FA20Ahtsham Shoukat: Aspect based Ranking of Online Product Reviews
  • MS FA20Ghulam Mehmood: Group Spam identification in Online Product Reviews
  • MS SP20Abid Ali: A Study on Diversification of Online Product Reviews

Final-year projects

  • FA19Spotting Spammer using Group Spammer Behaviour Analysis
  • FA19Detection and Diversification of Genuine Online Reviews
  • SP19Opinion Mining of Roman-Urdu reviews using Linguistic and Behaviour Analysis techniques
  • FA17Probabilistic Model for Opinion Spam Detection 1st of 53 FYPs
  • SP17Sentiment Analysis of Online Reviews Using Supervised Learning
  • SP16Election Prediction through Twitter Activity of Politicians, Voters and Journalists
B

Spatio-Temporal Data Mining

Spatio-temporal data relate to both space and time. Mining them discovers patterns in where and when people move: travel sequences from geotagged photos, points-of-interest recommendation, trajectory prediction, ride sharing and location privacy.

Selected publications

Theses

  • MS SP23Khurram Shahzad: A POI Clustered Users Recommendation Method in LBSNs considering the Weather Forecast
  • MS FA22Anam Khurshid: Predicting Tourist Destinations based on Interests and Travel Backgrounds using Text Analytics
  • MS SP22Syed Huzaifa Ansar: A Study of Crime Analysis and Prediction Using Data Mining Techniques
  • MS FA21Sajeel-UR-Rehman: Understanding User Mobility Behavior: Check-in Patterns in Location based Social Networks
  • MS FA21Umair Khalid: Long and Short Term User Preference Learning for Next POI Recommendation
  • MS FA20Umer Rana: Personal Route Prediction by Trajectory Data Mining
  • MS SP16Naeem Qasim: Preference oriented mining techniques for location based point search
  • MS SP14Maria Mushtaq: Developing Trust in Ride Sharing System
  • MS SP14Samia Arshad: A New Method of Optimization for Dynamic Ride Sharing System

Final-year projects

  • SP21Smart Travel Planner 1st of 96 FYPs
  • SP21Spatio-Temporal Data Mining based Task Reminder System
  • FA18Tourism Support System for Pakistan (co-supervised with Dr. Salman Khan)
  • FA15Intelli-Rem: A Location Based Intelligent Reminder for Smartphones
  • SP14Dynamic Ridesharing
C

Context Awareness & Behavior Mining

Context-aware computing uses situational and environmental information about people, places and things to anticipate needs and offer the right content and functions. We mine desktop and app interaction logs to understand user tasks and assist proactively.

Funded project: Activity Pooler, a system for proactive desktop assistance. PI Dr. Hamid Turab Mirza, HEC Pakistan, 2016–2017, PKR 0.39 million.

Selected publications

Theses

  • PhD 2012Hamid Turab Mirza: Research on Desktop Activity based on User Interaction Log Data Mining
  • PhDZishan Ali (in progress): Mining Behavior of Digital Users for Intelligent Task Assistance
  • MS SP19Zishan Ali: Periodic Mining of Users Routine Desktop Behavior for Task Assistance

Final-year projects

  • FA15Activity Based Desktop Search
  • FA15Desktop Activity Rhythmic Modelling (DARM)
  • SP15Interruption Handling on Computers in Workplace Environments
  • FA14Task Management System for Desktop Users
  • FA13Prototyping Tool & Task Management System
D

Mobility Assistance & Accessibility

Accessibility means technology usable by people with a wide range of abilities, following universal design so each user can interact in the way that suits them. Our work includes auditory feedback and sonification for visually impaired navigation and assistive apps.

Selected publications

See every publication →