Search the repository

A+ | A- | | SLO | ENG

Abstract: Introduction: Recognizing and extracting key information from textual data plays an important role in intelligent systems by maintaining up-to-date knowledge, reinforcing informed decision-making, question-answering, and more. It is especially apparent in the food domain, where critical information guides the decisions of nutritionists and clinicians. The information extraction process involves two natural language processing tasks named entity recognition—NER and named entity linking—NEL. With the emergence of large language models (LLMs), especially ChatGPT, many areas began incorporating its knowledge to reduce workloads or simplify tasks. In the field of food, however, we noticed an opportunity to involve ChatGPT in NER and NEL. Methods: To assess ChatGPT's capabilities, we have evaluated its two versions, ChatGPT-3.5 and ChatGPT-4, focusing on their performance across both NER and NEL tasks, emphasizing food-related data. To benchmark our results in the food domain, we also investigated its capabilities in a more broadly investigated biomedical domain. By evaluating its zero-shot capabilities, we were able to ascertain the strengths and weaknesses of the two versions of ChatGPT. Results: Despite being able to show promising results in NER compared to other models. When tasked with linking entities to their identifiers from semantic models ChatGPT's effectiveness falls drastically. Discussion: While the integration of ChatGPT holds potential across various fields, it is crucial to approach its use with caution, particularly in relying on its responses for critical decisions in food and bio-medicine.
Keywords: ChatGPT, food data, named-entity recognition, named-entity linking
Published in DiRROS: 16.09.2024; Views: 293; Downloads: 121
Full text (1,08 MB)
This document has many files! More...

Abstract: This study examines the generalization ability of algorithm performance prediction models across various bench-mark suites. Comparing the statistical similarity between the problem collections with the accuracy of performance prediction models that are based on exploratory landscape analysis features, we observe that there is a positive correlation between these two measures. Specifically, when the high-dimensional feature value distributions between training and testing suites lack statistical significance, the model tends to generalize well, in the sense that the testing errors are in the same range as the training errors. Two experiments validate these findings: one involving the standard benchmark suites, the BBOB and CEC collections, and another using five collections of affine combinations of BBOB problem instances.
Keywords: meta-learning, single-objective optimization, module importance
Published in DiRROS: 16.09.2024; Views: 315; Downloads: 126
Full text (1,29 MB)
This document has many files! More...

Keywords: bilevel optimization, evolutionary algorithms, pessimistic approach
Published in DiRROS: 16.09.2024; Views: 216; Downloads: 373
Full text (3,38 MB)
This document has many files! More...

Abstract: V končnem poročilu so predstavljeni rezultati terenskih raziskav šestih varstveno pomembnih vrst hroščev v Sloveniji v letih 2013 in 2014. Za močvirskega krešiča (Carabus variolosus), rogača (Lucanus cervus), alpskega kozlička (Rosalia alpina) in bukovega kozlička (Morimus funereus) je v Sloveniji že vzpostavljena shema populacijskega in distribucijskega monitoringa s pripadajočimi metodološkimi protokoli in je pričujoča naloga poročilo o stanju v letih 2013 in 2014 glede na podatke zbrane v okviru populacijskega in distribucijskega monitoringa. Za tri vrste (močvirski krešič, rogač in alpski kozliček) smo izračunali populacijske trende in ugotovili zmeren upad populacije močvirskega krešiča ter stabilni populaciji tako rogača kot alpskega kozlička. Za alpskega in bukovega kozlička smo po zaključku petletnega ciklusa snemanja evalvirali shemo monitoringa in podali predloge izboljšave sheme za nadaljnji monitoring vrst. V poročilu so predstavljeni rezultati popisa v letih 2013 in 2014 za dopolnitev razširjenosti vrste puščavnika (Osmoderma eremita) in škrlatnega kukuja (Cucujus cinnaberinus). Glavni namen pričujoče naloge je bilo tudi testiranje metode lova puščavnika v letu 2014. Poznavanje razširjenosti škrlatnega kukuja (Cucujus cinnaberinus) v zahodni Sloveniji je slabše, zato je bila izvedena dodatna raziskava razširjenosti. Glede na dosedanja vedenja o populacijah vseh že potrjenih vrst hroščev evropskega varstvenega pomena pri nas, podajamo nekaj smernic nujnih raziskav, ki bi jih bilo potrebno iz tega naslova izvesti v Sloveniji. Pri tem smo se osredotočili le na dejansko potrjene vrste.
Keywords: hrošči, monitoring, ekologija
Published in DiRROS: 16.09.2024; Views: 280; Downloads: 411
Full text (4,37 MB)
This document has many files! More...

Abstract: This report represents the result of the scientific and technical review of Commission Decision 2010/477/EU in relation to Descriptor 8. The review has been carried out by the EC JRC together with experts nominated by EU Member States, and has considered contributions from the GES Working Group in accordance with the roadmap set out in the MSFD implementation strategy (agreed on at the 11th CIS MSCG meeting). The report is one of a series of reports (review manuals) including Descriptor 1, 2, 5, 7, 8, 9, 10 that conclude phase 1 of the review process and, as agreed within the MSFD Common Implementation Strategy, are the basis for review phase 2, towards an eventual revision of the Commission Decision 2010/477/EU. The report presents the state of the technical discussions as of 30 April 2015 (document version 8.0: ComDecRev_D8_V8), as some discussions are ongoing, it does not contain agreed conclusions on all issues. The views expressed in the document do not necessarily represent the views of the European Commission.
Published in DiRROS: 16.09.2024; Views: 217; Downloads: 180
Full text (1,29 MB)
This document has many files! More...