A5085967592- Semantic Web and Ontologies
- Advanced Database Systems and Queries
- Data Management and Algorithms
- Water Quality Monitoring Technologies
- Energy Efficient Wireless Sensor Networks
- Service-Oriented Architecture and Web Services
- Logic, Reasoning, and Knowledge
- Web Data Mining and Analysis
- Natural Language Processing Techniques
- Recommender Systems and Techniques
- Data Mining Algorithms and Applications
- Complex Network Analysis Techniques
- Context-Aware Activity Recognition Systems
- Air Quality Monitoring and Forecasting
- IoT and Edge/Fog Computing
- Peer-to-Peer Network Technologies
- Misinformation and Its Impacts
- Integrated Energy Systems Optimization
- Digital Marketing and Social Media
- Network Security and Intrusion Detection
- Sentiment Analysis and Opinion Mining
- Advanced Graph Neural Networks
- Topic Modeling
- Data Quality and Management
- Underwater Vehicles and Communication Systems
University of Prishtina
2014-2024
University of Freiburg
2001-2019
During the pandemic, when people needed to physically distance, social media platforms have been one of outlets where expressed their opinions, thoughts, sentiments, and emotions regarding pandemic situation. The core object this research study is sentiment analysis peoples’ opinions on Facebook current situation in low-resource languages. To do this, we created a large-scale dataset comprising 10,742 manually classified comments Albanian language. Furthermore, paper report our efforts...
Receiving a recommendation for certain item or place to visit is now common experience. However, the issue of trustworthiness regarding recommended items/places remains one main concerns. In this paper, we present an implementation Naive Bayes classifier, most powerful classes Machine Learning and Artificial Intelligence algorithms in existence, improve accuracy raise confidence users items within network. Our approach proven as feasible one, since it reached prediction 89%, with...
New language technologies are coming, thanks to the huge and competing private investment fuelling rapid progress; we can either understand foresee their effects, or be taken by surprise spend our time trying catch up. This report scketches out some transformative new that likely fundamentally change use of language. Some these may feel unrealistically futuristic far-fetched, but a central purpose this - wider LITHME network is illustrate mostly just logical development maturation currently...
The internet of things (IoT) as an evolving technology represents active scientific research field in recognizing challenges associated with its application various domains, ranging from consumer convenience, smart energy, and resource saving to IoT enterprises. Sensors are crucial components that relay the collected data form stream for further processing. Interoperability connected digital resources is a key challenge environments. enrichment raw sensor semantic annotations using concept...
This dataset was collected as part of the InWaterSense project with a wireless sensor network (WSN) installed in site river Sitnica Kosovo, case study for monitoring remotely, continuously and real-time surface water quality Kosovo how it can be extended to all waters country assurance. Values four parameters are provided dataset, i.e., temperature, electrical conductivity, pH, dissolved oxygen measured by respective static sensors WSN time frame between May 2015 beginning January 2016 every...
Semantic technologies have been extensively used for integrating stream data applications. However, using SWRL, which has become the de facto standard rule language in Web, never Its open world assumption and monotonic nature makes SWRL powerless doing continuous inference over data. For example, aggregate functions on a particular window of streams cannot be expressed SWRL. C-SPARQL is framework supports querying streams. We introduce here C-SWRL, unified Web reasoning system that further...
The recent proliferation of Massive Open Online Courses (MOOCs) has initiated a plethora research endeavors revolving around new pedagogical methods in higher education. Integrating MOOCs blended learning can be beneficial different ways for both learners and instructors. In this position paper, we aim to provide brief comprehensive review about the challenges that education institutions Macedonia Kosovo face while coping with trends flexible or learning. Moreover, after describing some real...
Modeling co-authorship networks merely in FOAF, as is common with social networks, suffers the inability to capture semantics which are specific collaboration schemes scientific community. For instance, weight of co-author relation expresses magnitude that between two co-authors subject change dynamically new publications assigned graph. They require mechanisms like rules Semantic Web calculate them. We introduce here a framework (1) extending FOAF ontology, and (2) implementing SWRL...
Wireless sensor networks (WSNs) are aiding water quality monitoring with support for real-time and remote measurements in terrain. Environmental portals receiving data from sensors have been a practice since while among researchers. However, the Web portal introduced here is essentially an integrated it supports modeling management of both, observational stream on coming wireless – dynamic data, as well describing WSN itself, its devices corresponding site allocation static data. Access...
Accurate polyp detection during the colonoscopy procedure impacts colorectal cancer prevention and early detection. In this paper, we investigate influence of skip connections as main component encoder-decoder based convolutional neural network (CNN) architectures for We conduct experiments on long short further extend existing architecture by introducing dense lateral connections. The proposed segmentation utilizes in contracting path, moreover it between expanding path. Results obtained...
Sensors and other Internet of Things (IoT) technologies are increasingly finding application in various fields, such as air quality monitoring, weather alerts water healthcare etc. IoT sensors continuously generate large volumes observed stream data; therefore, processing requires a special approach. Extracting the contextual information essential for situational knowledge from sensor data is very difficult, especially when interpretation these required real time. This paper focuses on...
Association rules are used in recommender systems to develop a model that enhances the profiles of users, as well address cold start problem.Our approach proposes which is implemented system for recommending scientific papers called Collaborative Topic Regression (CTR).Collaborative consists two matrices, U and V, where represents relationship between users paper topics, while V topics.The CTR focused on matrix by adopting it, result, it influenced paper's textual content, leaving content...
A shift in water monitoring approach from traditional grab sampling to novel wireless sensors is gaining popularity not only among researchers but also the market. These latest technologies readily enable numerous advantageous arrangements like remote, continuous, real-time, and spatially-dense broad coverage measurements, identification of long-term trends parameters interest. Thus, a WSN system implemented river Kosovo as part InWaterSense project monitor its quality parameters. It one...
Nowadays, by help of tourist trip planners, in many travel destinations, the itinerary could be prepared automatically. The planners enable customization based on preferences, available time and budget, where is optimized using optimization techniques from field metaheuristics. In this paper, we present a new approach simulated annealing algorithm for solving planning problem. hard constraints include limited duration, working hours points interest (POIs) while preferences traveling comprise...
In this paper, we present an approach for classifying news articles as biased (i.e., hyperpartisan) or unbiased, based on a convolutional neural network. We experiment with various embedding methods (pretrained and trained the training dataset) variations of network architecture compare results. When evaluating our best performing actual test data set SemEval 2019 Task 4, obtained relatively low precision accuracy values, while gaining highest recall rate among all 42 participating teams.
The Internet of Things (IoT) vision is to connect uniquely identifiable devices that surround us the Internet, which best described through ontologies. Thereby, new emerging technologies such as wireless sensor networks (WSN) are recognized an essential enabling component IoT today. Hence, given increasing interest provide linked data Web either following Semantic Enablement (SWE) standard or Linked Data approach, there a need also explore those for potential hidden knowledge mining...
Association rule mining as descriptive data category aims to find interesting patterns on data. The quality of the is measured with two metrics: confidence and support. Especially in fields dealing sensitive data, such healthcare, resulting should be novel interesting. To achieve that, not only itself superior, but also other additional attributes added, do support results. That achieved by using Semantic Web technologies thus enriching used semantic relations between properties. A...
Stream data knowledge bases modelled with OWL are a proved natural approach. But, querying and reasoning over these is not supported standard Semantic Web technologies like SPARQL SWRL. Query processing systems enable querying, but to the best of our knowledge, rules still unable handle required features for effective inference stream data, i.e. non-monotonic, closed-world time-aware reasoning. In absence such system, we showed in previous work how Jess can be used monitoring water quality,...
Stream data knowledge bases modelled with OWL are a proved natural approach. But, querying and reasoning over these is not supported standard Semantic Web technologies like SPARQL SWRL. Query processing systems enable querying, but to the best of our knowledge, rules still unable handle required features for effective inference stream data, i.e. non-monotonic, closed-world time-aware reasoning. In absence such system, we showed in previous work how Jess can be used monitoring water quality,...