Dejan Lavbič

Ontology‐based multi‐agent system to support business users and management

Abstract For some decision processes a significant added value is achieved when enterprises’ internal Data Warehouse (DW) can be integrated and combined with external data gained from web sites of competitors and other relevant Web sources. In this paper we discuss the agent‐based integration approach using ontologies (DSS‐MAS). In this approach data from internal DW and external sources are scanned by coordinated group of agents, while semantically integrated and relevant data is reported to business users according to business rules. After data from internal DW, Web sources and business rules are acquired, agents using these data and rules can infer new knowledge and therefore facilitate decision making process. Knowledge represented in enterprises’ ontologies is acquired from business users without extensive technical knowledge using user friendly user interface based on constraints and predefined templates. The approach presented in the paper was verified using the case study from the domain of mobile...

Collective information extraction using first-order probabilistic models

Traditional information extraction (IE) tasks roughly consist of named-entity recognition, relation extraction and coreference resolution. Much work in this area focuses primarily on separate subtasks where best performance can be achieved only on specialized domains. In this paper we present a collective IE approach combining all three tasks by employing linear-chain conditional random fields. The usage of probabilistic models enables us to easily communicate between tasks on the fly and error correction during the iterative process execution. We introduce a novel iterative-based IE system architecture with additional semantic and collective feature functions. Proposed system is evaluated against real-world data set, introduced in the paper, and results are better over traditional approaches on two tested tasks by error reduction and performance improvements.

Rule- and context-based dynamic business process modelling and simulation

Six requirements of dynamic business process (DBP) is defined.In DBP each next activity is selected according to predefined rules and context.A rule- and context-based DBP modelling and simulation method is proposed.A reference architecture for a DBP simulation tool is developed.The proposed architecture was implemented into DBP simulation prototype. The traditional approach used to implement a business process (BP) in todays information systems (IS) no longer covers the actual needs of the dynamically changing business. Therefore, a necessity for a new approach of dynamic business process (DBP) modelling and simulation has arisen. To date, existing approaches to DBP modelling and simulation have been incomplete, i.e. they lack theory or a case study or both. Furthermore, there is no commonly accepted definition of BDP. Current BP modelling tools are suitable almost solely for the modelling and simulation of a static BP that strictly prescribes which activities, and in which sequence, to execute. Usually, a DBP is not defined strictly at the beginning of its execution, and it changes under new conditions at runtime. In our paper, we propose six requirements of DBP and an approach for rule- and context-based DBP modelling and simulation. The approach is based on changing BP rules, BP actions and their sequences at process instance runtime, according to the new business system context. Based on the proposed approach, a reference architecture and prototype of a DBP simulation tool were developed. Modelling and simulation were carried out using this prototype, and the case study shows correspondence to the needs of dynamically changing business, as well as possibilities for modelling and simulating DBP.

Agent Oriented Approach for Integration of BI Systems

Decision support in enterprises is gaining its importance in the age of Internet and e-business. While dealing with the need for fast response in the dynamic competitive environments and rapid increase of information available on various networks we propose multi-agent system approach to backend decision support. Research case study presented in the paper is from the domain of mobile communications with systems main goal of delivering right information at the right time to the right users. We integrate several BI systems such as data mining and data warehousing with information available on the Internet. Ontologies are used to store derived knowledge, to support knowledge exchange between our MAS and other systems, to support agent-to-agent communication and enable seamless extending of MAS capabilities.

What Language Do Stocks Speak

Stock prediction is a challenging and chaotic research area where many variables are included with their effects being complex to determine. Nevertheless, stock value prediction is still very appealing for researchers and investors since it might be profitable, yet the number of published research papers remains to be relatively small. The employment of advanced data analysis techniques has already been suggested by previous researches, such as the use of neural networks for stock price prediction, but practical implications of the majority of approaches are limited as they are concerned mainly with a prediction accuracy and less with the success in real trading with consideration of trading fees. We propose a novel approach for stock trend prediction that combines Japanese candlesticks (OHLC trading data) and neural network based group of models Word2Vec. Word2Vec is usually utilized to produce word embeddings in natural language processing tasks, while we adopt it for acquiring semantic context of words in candlesticks’ sequence, where clustered candlesticks represent stock’s words. The approach is employed for the extraction of useful information from large sets of OHLC trading data to improve prediction accuracy. In evaluation of our approach we define a trading strategy and compare our approach with other popular prediction models – Buy & Hold, MA and MACD. The evaluation results on Russell Top 50 index are encouraging – the proposed Word2Vec approach outperformed all compared models on a test set with a statistical significance.

LogMap+: Relational data enrichment and linked data resources matching

Relational database to ontology mapping and ontology matching techniques are mostly addressed separately, even though it is known that the real power of semantic data lies in data interconnection. The latter is especially important when designing a new ontology, which often includes at least some of the concepts that already exist in the linked open data cloud. Thus, in this paper we describe a new end-to-end tool LogMap+ for transformation of relational data into an ontology and matching it against a pre-existent semantic source. Apart from offering the efficient web-based application, the main contributions are the improvements of the domain specific LogMap system. We evaluate our general tool against OAEI 2014 challenge datasets and achieve comparable results to the top performing algorithms and also outperform the domain specific LogMap tool.

Improving information quality of Wikipedia articles with cooperative principle

Purpose The purpose of this paper is to investigate the impact of cooperative principle on the information quality (IQ) by making objects more relevant for consumer needs, in particular case Wikipedia articles for students. Design/methodology/approach The authors performed a quantitative study with participants being invited to complete an online survey. Each rater evaluated three selected and re-written articles from Wikipedia by four IQ dimensions (accuracy, completeness, objectivity, and representation). Grice’s maxims and submaxims were used to re-write articles and make them more relevant for student cognitive needs. The results were analyzed with statistical methods of mean, standard deviation, Cronbach’s α, and ICC (two-way random model of single measure). Findings The study demonstrates that Wikipedia articles can be made more relevant for student needs by using cooperative principle with increase in IQ and also achieving higher consistency of students’ scores as recent research. In particular, students in the research perceived the abstract, constructed with cooperative principle, more objective and complete as reported in recent research. Practical implications The work can benefit encyclopedia editors to improve IQ of existing articles as well as consumers that would obtain more relevant information in less reading time. Originality/value This is one of the first attempts to empirically investigate the application of cooperate principle to make objects more relevant for consumer needs and impact of this on IQ. IQ improvement evidence is provided and impacts on IQ dimensions such as objectivity, completeness, accuracy, and representation for research community to validate and compare results.

Recommender system for learning SQL using hints

ABSTRACT Today’s software industry requires individuals who are proficient in as many programming languages as possible. Structured query language (SQL), as an adopted standard, is no exception, as it is the most widely used query language to retrieve and manipulate data. However, the process of learning SQL turns out to be challenging. The need for a computer-aided solution to help users learn SQL and improve their proficiency is vital. In this study, we present a new approach to help users conceptualize basic building blocks of the language faster and more efficiently. The adaptive design of the proposed approach aids users in learning SQL by supporting their own path to the solution and employing successful previous attempts, while not enforcing the ideal solution provided by the instructor. Furthermore, we perform an empirical evaluation with 93 participants and demonstrate that the employment of hints is successful, being especially beneficial for users with lower prior knowledge.

The Role of Social Connections in Plagiarism Detection

Plagiarism is considered as an unethical act. Over the past few years its rate has increased considerably due to a widespread access to electronic documents on the Web. Existing tools for plagiarism detection are not efficient enough and if we want to successfully prevent these kind of acts we must improve today’s plagiarism detection approaches. The paper proposes a framework for improved detection of plagiarism, where we focus on integration of information from social networks, information from the Web and semantically enriched visualization of information about authors and plagiates. Visualization enables exploring data and seeking of advanced patterns of plagiarism. We also developed a special tool to support the proposed framework. The results of evaluation confirmed our hypothesis that employment of social network analysis and advanced visualization techniques improves plagiarism detection process.

Rapid Development of Executable Ontology for Financial Instruments and Trading Strategies

In this paper we employ Rapid Ontology Development approach (ROD) with constant evaluation of steps in the process of ontology construction for development of Financial Instruments and Trading Strategies (FITS) ontology. We show that ontology development process does not conclude with successful definition of schematic part of ontology but we continue with post development activities where additional axiomatic information and instances with dynamic imports from various sources are defined. The result is executable ontology as part of Semantic Web application that uses data from several semi structured sources. The overall process of construction is suitable for users without extensive technical and programming skills and those users are rather experts in the problem domain.