
Lavdim HalilajBosch · Department of Corporate Research
Lavdim Halilaj
PhD
About
50
Publications
39,573
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
584
Citations
Introduction
Additional affiliations
April 2019 - April 2022
May 2010 - May 2013
Kosovo Energy Corporation
Position
- Developer
December 2008 - May 2010
Asseco Group
Position
- Developer
Publications
Publications (50)
Collaborative vocabulary development in the context of data integration is the process of finding consensus between the experts of the different systems and domains. The complexity of this process is increased with the number of involved people, the variety of the systems to be integrated and the dynamics of their domain. In this paper we advocate...
Vocabularies are increasingly being developed on platforms for hosting version-controlled repositories, such as GitHub. However, these platforms lack important features that have proven useful in vocabulary development. We present VoCol, an integrated environment that supports the development of vocabularies using Version Control Systems. VoCol is...
Making an informed and right decision poses huge challenges for drivers in day-to-day traffic situations. This task vastly depends on many subjective and objective factors, including the current driver state, her destination, personal preferences and abilities as well as surrounding environment. In this paper, we present CoSI (Context and Situation...
The information perceived via visual observations of real-world phenomena is unstructured and complex. Computer vision (CV) is the field of research that attempts to make use of that information. Recent approaches of CV utilize deep learning (DL) methods as they perform quite well if training and testing domains follow the same underlying data dist...
Traditional approaches to computer vision tasks based on neural networks are typically trained on a large amount of pure image data. By minimizing the cross-entropy loss between a prediction and a given target class, the network and its visual embedding space are automatically learned to fullfill a given task. However, due to the sole dependence on...
Despite the remarkable success of deep neural networks (DNNs) in computer vision, they fail to remain high-performing when facing distribution shifts between training and testing data. In this paper, we propose Knowledge-Guided Visual representation learning (KGV), a distribution-based learning approach leveraging multi-modal prior knowledge, to im...
Knowledge graphs (KGs) have transformed data management within the manufacturing industry, offering effective means for integrating disparate data sources through shared and structured conceptual schemas. However, harnessing the power of KGs can be daunting for non-experts, as it often requires formulating complex SPARQL queries to retrieve specifi...
Trajectory prediction in autonomous driving relies on accurate representation of all relevant contexts of the driving scene, including traffic participants, road topology, traffic signs, as well as their semantic relations to each other. Despite increased attention to this issue, most approaches in trajectory prediction do not consider all of these...
Trajectory prediction in traffic scenes involves accurately forecasting the behaviour of surrounding vehicles. To achieve this objective it is crucial to consider contextual information, including the driving path of vehicles, road topology, lane dividers, and traffic rules. Although studies demonstrated the potential of leveraging heterogeneous co...
Precisely predicting the future trajectories of surrounding traffic participants is a crucial but challenging problem in autonomous driving, due to complex interactions between traffic agents, map context and traffic rules. Vector-based approaches have recently shown to achieve among the best performances on trajectory prediction benchmarks. These...
Motion prediction and planning are key components to enable autonomous driving. Although high definition (HD) maps provide important contextual information that constrains the action space of traffic participants, most approaches are not able to fully exploit this heterogeneous information. In this work, we enrich the existing road geometry of the...
Current search engines are heavily optimized and excel on retrieving information based on a given set of keywords. The more sophisticated ones are extended to support searches based on the sentences or full text by calculation the similarity of the given query with the already stored entries. However, accessing the domain information using natural...
The usage of semantic technologies for purposes like data integration, information retrieval, search, and decision-making is steadily on the rise. Bosch is
leveraging these technologies to enhance the ability to represent, integrate, and
query various data sources. The main objectives are enabling data understanding, interlinking, and analysis. To...
Autonomous Driving (AD) datasets, when used in combination with deep learning techniques, have enabled significant progress on difficult AD tasks such as perception, trajectory prediction and motion planning. These datasets represent the content of driving
scenes as captured by various sensors, including cameras, RADAR, and LiDAR, along with 2D/3D...
Current deep learning methods for object recognition are purely data-driven and require a large number of training samples to achieve good results. Due to their sole dependence on image data, these methods tend to fail when confronted with new environments where even small deviations occur. Human perception, however, has proven to be significantly...
Representing relevant information of a traffic scene and understanding its environment is crucial for the success of autonomous driving. Modeling the surrounding of an autonomous car using semantic relations, i.e., how different traffic participants relate in the context of traffic rule based behaviors, is hardly been considered in previous work. T...
Automated driving is one of the most active research areas in computer science. Deep learning methods have made remarkable breakthroughs in machine learning in general and in automated driving (AD) in particular. However, there are still unsolved problems to guarantee reliability and safety of automated systems, especially to effectively incorporat...
Automated Driving (AD) datasets, when used in combination with deep learning techniques, have enabled significant progress on difficult AD tasks such as perception, trajectory prediction and motion planning. These datasets represent the content of driving scenes as captured by various sensors, including cameras, RADAR, and LiDAR, along with 2D/3D a...
Learned latent vector representations are key to the success of many recommender systems in recent years. However, traditional approaches like matrix factorization produce vector representations that capture global distributions of a static recommendation scenario only. Such latent user or item representations do not capture background knowledge an...
Knowledge graph embeddings (KGE) are vector representations that capture the global distributional semantics of each entity instance and relation type in a static Knowledge Graph (KG). While KGEs have the capability to embed information related to an entity into a single representation, they are not customizable to a specific context. This is funda...
The task of safe driving poses a huge challenge for drivers in day to day driving situations. Many times, this task can be very difficult, e.g., due to dense traffic, bad weather conditions, or a risky driving manoeuvrer, and thus demand high concentration of the driver. The difficulty level escalates by the ever-increasing infotainment offers insi...
Context-aware Recommender Systems (CARS) are becoming an integral part of the everyday life by providing users the ability to retrieve relevant information based on their contextual situation. To increase the predictive power considering many parameters, such as mood, hunger level and user preferences, information from heterogeneous sources should...
Deep learning techniques achieve high accuracy in computer vision tasks. However, their accuracy suffers considerably when they face a domain change, i.e., as soon as they are used in a domain that differs from their training domain. For example, a road sign recognition model trained to recognize road signs in Germany performs poorly in countries w...
Over the years, the demand for interoperability support between diverse applications has significantly in-creased. The Resource Definition Framework (RDF), among other solutions, is utilized as a data modelinglanguage which allows for encoding the knowledge from various domains in a unified representation. More-over, a vast amount of data from hete...
The era of digitalization poses high demands on capturing and processing knowledge generated in everyday life in formal models.
Ontologies provide common means for formal knowledge capturing and modelling for a universe of discourse. Developing ontologies, however, can be a complex, time-consuming and expensive process which requires a significant...
Cyber-Physical Systems (CPS) are engineered systems that result from the integration of both physical and computational components designed from different engineering perspectives (e.g., mechanical, electrical, and software). Standards related to Smart Manufacturing (e.g., AutomationML) are commonly used to describe CPS components, as
well as to fa...
Budget and spending data are among the most published Open Data datasets on the Web and continuously increasing in terms of volume over time. These datasets tend to be published in large tabular files – without predefined standards – and require complex domain and technical expertise to be used in real-world scenarios. Therefore, the potential bene...
The development of domain-specific ontologies requires joint efforts among different groups of stakeholders, such as knowledge engineers and domain experts. During the development processes, ontology changes need to be tracked and propagated across developers. Version Control Systems (VCSs) collect metadata describing changes and allow for the sync...
The development of domain-specific ontologies requires joint efforts among different groups of stakeholders, such as ontology engineers and domain experts. By following a test-driven development technique, a set of test cases ensures that ontology changes do not violate predefined requirements. However, since the number of test cases can be large a...
Cyber-Physical Systems (CPS) are engineering systems that result from the integration of both physical and computational components designed from different engineering perspectives. Standards addressing smart manufacturing, such as AutomationML, can be used to describe CPS components and to facilitate their integration. Albeit expressive, these sta...
Cyber-Physical Systems (CPS) are engineering systems that result from the integration of both physical and computational components designed from different engineering perspectives. Standards addressing smart manufacturing, such as AutomationML, can be used to describe CPS components and to facilitate their integration. Albeit expressive, these sta...
The digitization of the industry requires information models describing assets and information sources of companies to enable the semantic integration and interoperable exchange of data. We report on a case study in which we realized such an information model for a global manufacturing company using semantic technologies. The information model is c...
Interoperability among actors, sensors, and heterogeneous systems is a crucial factor for realizing the Industry 4.0 vision, i.e., the creation of Smart Factories by enabling intelligent human-to-machine and machine-to-machine cooperation. In order to empower interoperability in Smart Factories, standards and reference architectures have been propo...
Traditional approaches, which follow a test-driven development technique, allow a set of test cases to be exhaustively evaluated ensuring that each modification of an ontology does not violate predefined requirements. However, the time required for the evaluation of test cases is high and usually represents a bottleneck in an ontology development p...
OpenBudgets.eu (OBEU) provides an open-source software framework and accompanying Software-As-A-Service (SAAS) platform for supporting financial transparency, thus enhancing accountability within public sectors as well as influencing corruption prevention. To this end, a scalable framework for multi-stakeholders is developed, with the aim of maximi...
Vocabularies are increasingly being developed on platforms for hosting version-controlled repositories, such as GitHub. However, these platforms lack important features that have proven useful in vocabulary development. We present VoCol, an integrated environment that supports the development of vocabularies using Version Control Systems . VoCol is...
A Version Control System (VCS) is usually required for successful ontology development in distributed settings. VCSs enable tracking and propagation of ontology changes, as well as collecting metadata to describe changes, e.g., who made a change at which point in time. Modern VCSs implement an optimistic approach that allows for simultaneous change...
We present our work towards a smart, community and crowd driven data repository in the context of the SDI-X project. The goal is to extend the Smart Data Innovation Lab (SDIL) towards a platform offering an extensive collection of Big Data datasets from various areas and sources of industrial and academic research as well as state-of-the-art hardwa...
Industry 4.0 standards, such as AutomationML, are used to specify properties of mechatronic elements in terms of views, such as electrical and mechanical views of a motor engine. These views have to be integrated in order to obtain a complete model of the artifact. Currently, the integration requires user knowledge to manually identify elements in...
Unification and automation of RESTful web ser-vices' documentation and descriptions is currently receiving increasing attention. The open-source OpenAPI Specification (formerly known as Swagger) has become core of this effort and has been adopted by a number of major companies. It allows the description of RESTful web services using objects represe...
Collaborative vocabulary development in the context of data integration is the process
of finding consensus between experts with different backgrounds, system understanding
and domain knowledge. The complexity of this process increases with the number of
people involved, the variety of the systems to be integrated and the dynamics of their
domain....
In the engineering and manufacturing domain, there is currently an atmosphere of departure to a new era of digitized production. In different regions, initiatives in these directions are known under different names, such as industrie du futur in France, industrial internet in the US or Industrie 4.0 in Germany. While the vision of digitizing produc...
A major bottleneck for a wider deployment and use of ontologies and knowledge engineering techniques is the lack of established conventions along with cumbersome and inefficient support for vocabulary and ontology authoring. We argue, that the pragmatic development by convention paradigm well-accepted within software engineering, can be successfull...
Vocabularies typically reflect a consensus among experts in a
certain application domain. They are thus implemented in collaboration
of domain experts and knowledge engineers. Particularly the presence of
domain experts with little technical background requires a low-threshold
vocabulary engineering methodology. This methodology should be im-
pleme...
Author ranking is growing in popularity since search engines are considering the author’s reputation of a Web page when generating search results. A question that naturally arises is whether we should rank authors on the Web as we rank Web pages by considering their links. In addition, over what links to actually calculate author ranking? We have a...