Adrian Groza

Compliance checking of integrated business processes

To enable compliance checking on integrated business processes we developed the NTL-ALC logical framework, for closing the gap between the abstract norms and the concrete business processes. To reason on the active obligations and permissions, we extended the normative temporal logic NTL, by applying the deontic operators O (obligation) and P (permission) on concepts of the ALC (Attribute Language with Complements) description logic. As proof of concept of our results we have used the Hazard Analysis at Critical Control Points (HACCP) standard, aiming to prevent the occurrence of significant hazards in the food industry.

Arguing with justifications between collaborating agents

We exploit the Justification Logic capabilities of reasoning about justifications, comparing pieces of evidence, and measuring the complexity of justifications in the context of argumentative agents. The research can be integrated into the larger context of integrating logic and argumentation. The paper introduces distributed justification logic

Integrating DBpedia and SentiWordNet for a tourism recommender system

\mathcal{DJL}

Improving remote sensing crop classification by argumentation-based conflict resolution in ensemble learning

as an extension of justification logic for multi-agent systems, and it also investigates the expressivity of

Running contracts with defeasible commitment

\mathcal{DJL}

LELA - a natural language processing system for Romanian tourism

for argumentative agents. Not knowing all of the implications of their knowledge base, agents use justified arguments for reflection and guidance.

Plausible Description Logic Programs for Stream Reasoning

The popularity of the social web introduces opportunities for the recommender systems, whilst new challenges arise when semantic knowledge is integrated in the landscape. The large amount of opinions available from Web 2.0 are exploited here to improve recommendation techniques in a semantic context. The developed recommendation system matches the crawled opinions against tourist objectives within the DBpedia ontology. Following a natural language processing step in Gate, several metrics are employed to build a recommendation plan, and formal justification is provided in case of need.

Mining traffic patterns from public transportation GPS data

A hybrid system for crop classification from satellite images is proposed.A novel method for conflict resolution in ensemble learning is developed.Argumentation technology performs dialectical analysis on debatable instances.Agricultural expert knowledge is merged with rules extracting from base learners.Classification accuracy and transparency of the decision are increased. The acquisition of data through remote sensing has become of great importance in precision agriculture, as it covers large geographical areas faster and cheaper than ground inspections. The challenge is to develop technical solutions that can benefit from both huge amounts of raw data extracted from satellite images, but also from the robust amount of knowledge refined during centuries of agricultural practice. Aiming to accurately classify crops from satellite images, we developed a hybrid intelligent system that can exploit both agricultural expert knowledge and machine learning algorithms. As the crop raw data is characterized by heterogeneity, we drive our attention to ensemble learners, while expert knowledge is encapsulated within a rule-based system. Vote-based methods for solving conflicts between ensembles base learners have difficulties in classifying exceptional cases correctly and also to give the rationale behind their decision. The conceptual research question is on conflict resolution in ensemble learning. To deal with debatable cases in ensemble learning and to increase transparency in such debatable decisions, our hypothesis is that argumentation could be more effective than voting-based methods. The main contribution is that voting system in ensemble learning is substituted by an argumentation-base conflict resolutor. Prospective decisions of base classifiers are presented to an argumentative system based on defeasible logic that performs dialectical reasoning on pros and cons against a classification decision. The system computes a recommendation considering both the rules extracted from base learners and the available expert knowledge. The investigated case study deals with crop classification into four classes: corn, soybean, cotton, and rice. The test site used for the experiment is an area of 20 square kilometers in the New Madrid County, southeast of the Missouri State, USA. The results show that our approach increases classification accuracy compared to the voting-based method for conflict resolution in an ensemble learner comprising of three base classifiers: a decision tree, a neural network, and a support vector machine algorithm. We also argue that combining ensemble learning and argumentation fits the decision patterns of human agents, who first collect various opinions and then perform dialectical reasoning on these opinions. We think that the people who can benefit from the conceptual instrumentation presented in this work are decision makers in domains characterized by high data availability, robust expert knowledge, and a need for justifying the rationale behind decisions.

Automating the dispute resolution in a task dependency network

Real life contracts imply commitments which are active during their running window, with effects on both normal runs as well as in the case of exceptions. We have defined defeasible commitment machines (DCMs) to provide more flexibility. As an extension to the task dependency model for the supply chain we propose the commitment dependency network (CDN) to monitor contracts between members of the supply chain. The workings of the DCMs in the CDN is shown by a simple scenario with supplier, producer, and consumer.

Assuring safety in air traffic control systems with argumentation and model checking

This paper presents a commercial semantic-based system for the Romanian tourism. The Lela system exploits both open linked data from Romanian and international sources, and also proprietary databases in the tourism domain. We present the process of creating the linked data set, based on: i) engineering the LELA Romanian tourism ontology, and ii) populating the ontology by linking open data. The system also provides a natural language interface for the Romanian language. The queries are automatically translated into SPARQL based on a controlled vocabulary derived from the Lela ontology.