- Scientific Computing and Data Management
- Distributed and Parallel Computing Systems
- Semantic Web and Ontologies
- Data Mining Algorithms and Applications
- Artificial Intelligence in Healthcare
- Advanced Database Systems and Queries
- AI in cancer detection
- Mobile Learning in Education
- Advanced Clustering Algorithms Research
- Data Management and Algorithms
- Radiomics and Machine Learning in Medical Imaging
- Imbalanced Data Classification Techniques
- Sentiment Analysis and Opinion Mining
- Simulation Techniques and Applications
- Soil and Land Suitability Analysis
- Data-Driven Disease Surveillance
- Online and Blended Learning
- Technology Adoption and User Behaviour
- Data Analysis with R
- Educational and Technological Research
- Innovative Teaching Methods
- Text and Document Classification Technologies
- Advanced Text Analysis Techniques
- Rough Sets and Fuzzy Logic
- Research Data Management Practices
Sukkur IBA University
2016-2023
University of Bayreuth
2007-2009
Abstract The k-means, one of the most widely used clustering algorithm, is not only faster in computation but also produces comparatively better clusters. However, it has two major downsides, first sensitive to initialize k value and secondly, especially for larger datasets, number iterations could be very large, making computationally hard. In order address these issues, we proposed a scalable cost-effective called R-k-means, which provides an optimized solution large scale high-dimensional...
Recently; medical data mining has become one of the well-established research areas machine learning and AI base techniques have been used to solve complex classification problem thyroid disease. Due existence non-palpable nodules it is very hard detect structural changes disease by assessing functional changes. For instance at level "Euthyroid" normal hormonal state but this would be involved in initial such as goiter, cold nodule, MNG (multiple nodule goiter) cancer (Grave's Disease so...
In the process of knowledge discovery, reliability results depends upon effectiveness attributes selected for decision. The curse dimensionality refers to phenomenon in which excessive number dimensions affect analysis. order eradicate text analysis, we are proposing an ontology-based semantic measure intelligent selection/reduction features. Among various mining techniques, has a significant contribution field. measures, mathematical models used find similarity between concepts ontology,...
Recently; AI based methods are frequently used in healthcare industry to unfold historical hindsight explore the insight and envisage foresight. For example, identification of epidemiological patterns thyroid disease targeted area(s) supports stakeholders (government agencies, health organizations, NGOs, policy makers so on) formulating proper policies combat such kind fatal diseases. Also, predictive Future Visualization (FV) prevalence is really helpful for these properly focus on specific...
Actionable Knowledge Discovery approaches to extract the business and technical significant actions/patterns support direct decision making.These actions suggest how transform an object from undesirable status a desirable by incurring less cost high profit.This article aims propose work that generates actionable patterns efficiently.It reduces search space number of iterations for attribute value change during action generation.Performance proposed method is compared with Yang's OF-CEAMA on...
Learning from everywhere, at whatever time is a modern trend in the field of learning that supposed to be tactile with both cost and saving. This apparent way ICT technology intercede knowledge processes between students. There are studies literature draw attention disadvantages being utilized process learning. article presents SWOT analytic study performed on ubiquitous computing (u-computing) computer-intermediated community interaction its effects education. Interviews have been conducted...
In recent years, scientists are dealing more and with data intensive complex applications. Many scientific workflow systems emerged which adapt technology methods stemming from the management area that should support in understanding working their scenarios. However as these often descend problem solving environments, many of them missing a well structured conceptual method for process modeling execution foundation. this publication we present comprehensive developing analyzing based This is...
Data exchange between multiple sources in scientific applications poses significant data management issues which involve the transportation of from one system to another as well syntactic and semantic integration data, i.e. come different formats have meanings. In order deal with these a systematic structured way, we propose sophisticated framework based on process modeling.
In recent, medical Image mining has witnessed to be one of the emerging fields machine learning. Particularly; classification problem DICOM (Digital Imaging and Communications in Medicine) images become a prominent challenge. Thyroid cancer must detected as earlier possible; little delay would extremely proved hazards for human health may resulted into most fatal threat life. Infect in-depth study physical components cells FNAB (Fine needle aspiration Biopsy) help refine results provide more...
Due to the large volume and high complexity of data, end-users are often confronted with data management issues such as syntactic semantic integration (data comes in different formats has meanings) well pure movement between information systems. In order cope these a systematic structured manner, we propose an elaborate framework based on process modeling, provision, repository which tracks all central components our approach. It is out scope this paper reflect detail, instead main focus...
The research reveals that bank's customers prefer internet banking services over branch due to safety, security, convenience, cost-effectiveness, reliability, error-free system, and speed user-friendly. However, inaccessibility of ATM machine issues highly influences the selection choice Internet Banking. This study will help banks how they can increase availability ATMs what are potential ought be tackled improve customer's feedback. These include misreporting cash availability, out order,...
Due to the proliferation of data generating devices such as sensors in scientific applications, integration has become most challenging task since stemming from these are extremely heterogeneous terms structure (schema) and semantics (interpretation). In practice, transformation is typically performed by scientists manually; fact extensive efforts required. The approaches for automating much possible badly needed. DaltOn a generic framework that offers various functionalities managing...
Due to the high level exposure of biomedical image analysis, Medical mining has become one well-established research area(s) machine learning. AI (Artificial Intelligence) techniques have been vastly used solve complex classification problems thyroid cancer. Since persistence copycat chromatin properties and unavailability nuclei measurement techniques, it is really problem for doctors determine initial phases enlargement assess early changes distribution. For example involvement multiple...
Analysing the Education infrastructure has become a crucial activity in imparting quality teaching and resources to students. Facilitations required improving current education status future schools is an important analytical component. This best achieved through Geographical Information System (GIS) analysis of spatial distribution schools. In this work, we will execute GIS Analytics on rural urban school distributions Sindh, Pakistan. Using reliable dataset collected from international...