Sergejs Kodors

Translating knowledge into formal representation for the purpose of building an expert system is a daunting task for domain experts and requires information technology (IT) competence and software developer support. The availability of open and robust expert system shells is a way to solve this task. A new architecture of a rule-based expert system combining the digital twin paradigm and a capability-driven approach is presented in this study. The aim of the architecture is to provide a user-friendly framework for domain experts to build upon without the need to delve into technical aspects. To support this architecture, an open-source no-coding shell RecGen has been developed (Python and Django framework). RecGen was validated on a use case of an expert system for providing recommendations to reduce plate waste in schools. In addition, the article presents experiments with large language models (LLMs) by implementing a question-answering functionality in an attempt to improve the user experience while working with large expert system knowledge bases. A mean classification accuracy of 74.1% was achieved experimentally using the injection method with language prefixes. The ablation test was applied in order to investigate the effect of augmentation, injection, a linear layer size, and lowercase text on LLM accuracy. However, the analysis of the results showed that clustering algorithms would be a more suitable solution for future improvements of the expert system shell RecGen.

Flowering intensity is an important parameter to predict and control fruit yield. However, its estimation is often based on subjective evaluations of fruit growers. This study explores the application of the YOLO framework for flowering intensity estimation. YOLO is a popular computer vision solution for object-detecting tasks. It was applied to detect flowers in different studies. Still, it requires manual annotation of photographs of flowering trees, which is a complex and time-consuming process. It is hard to distinguish individual flowers in photos due to their overlapping and indistinct outlines, false positive flowers in the background, and the density of flowers in panicles. Our experiment shows that the small dataset of images (320 × 320 px) is sufficient to achieve an accuracy of 0.995 and 0.994 mAP@50 for YOLOv9m and YOLOv11m using aggregated mosaic augmentation. The AI-based method was compared with the manual method (flowering intensity estimation, 0–9 scale). The comparison was completed using data analysis and the MobileNetV2 classifier as an evaluation model. The analysis shows that the AI-based method is more effective than the manual method.

A fundamental step to foster a sustainable future is enhancing students’ awareness of responsible food consumption. The present research study assessed students’ awareness of food waste (FW) issues, attitudes towards school catering and lunch management, and the reasons for plate waste (PW) in Rezekne city schools, Latvia. A survey was conducted in April 2024 involving 944 students in grades 2 to 7 across four selected schools (S1, S2, S3, and S4), provided with state and municipality-funded free lunches. Statistical analysis methods (Mean ± SD, Kruskal–Wallis H test, Mann–Whitney U post hoc test, and Spearman’s correlation analysis) were used to analyze the results. The findings revealed limited awareness and knowledge of students about FW issues, and generally indifferent attitude towards FW and school meals. In addition, significant dissatisfaction with the sensory qualities of meals was observed, yet sensory satisfaction did not correlate with FW levels, pointing to broader issues within the catering model. The restrictive pre-served portion system was found ineffective in reducing FW and fostering responsible consumption. A sustainable catering model should empower students to independently choose the type and quantity of food, combining their abilities and motivation with practical actions. These findings provide a basis for strategies aimed at reducing FW and promoting responsible and sustainable food consumption in Latvian schools.

Food waste (FW) threatens food security, environmental sustainability, and economic efficiency, with about one-third of global food production lost or wasted. Schools play a crucial role in addressing FW, representing lost resources and missed educational opportunities. The present research assessed three interventions to reduce plate waste (PW) in Rezekne City schools, namely (S1) a plate waste tracker, (S2) an awareness and educational campaign, and (S3) organizational changes, including larger plates, extended lunch breaks, and teacher supervision. Implemented in three schools with a fourth as a control, PW was measured at three intervals, at pre-intervention, short-term, and long-term post-intervention. The PW data analysis utilized two models (day view and class view) and a Wilcoxon signed-rank test. While the plate waste tracker initially reduced PW, sustained impact required continuous reinforcement. The awareness and educational campaign alone proved insufficient, highlighting the need for complex strategies. The organizational changes unexpectedly increased PW, underscoring FW’s complexity. The research has concluded that reducing FW requires tailored and multi-faceted approaches. According to the MOA framework, the school catering model in Rezekne City lacks essential “Opportunities” for effective FW reduction, as students have limited flexibility in portion sizes and food choices, which hinders the interventions’ effectiveness. Future research should explore adaptable FW-reducing interventions suited to specific school contexts.

In the context of precision horticulture, decision support tools play a significant role in providing fruit growers with insights into orchard conditions, facilitating informed decisions regarding orchard management practices. This study presents the development of an autonomous yield estimation system designed to provide decision support to small commercial orchards. Autonomous yield estimation is based on the application of UAVs and AI. AI is used to identify and quantify fruitlets and fruits in photographs collected by UAV. In this article, we present our prototype of an autonomous yield estimation system. The adapted “4+1” architecture was applied to design a system with a holistic approach analyzing software, hardware, and ecosystem requirements. Six datasets are presented, which contain the images of fruitlets and fruits of apples, pears, and cherries. Three CNN models were trained: YOLOv8m, YOLOv9m, and YOLOv10m. The experiment showed that the most accurate was YOLOv9m, which achieved mean accuracies of 0.896 mAP@50 and 0.510 mAP@50:95 for all datasets.

Food waste indicates ineffective and irresponsible consumption of resources, particularly during the food consumption stage. The aim of our research study is to optimize the catering management process at Latvian schools by reducing the amount of plate waste. The experts developed a set of recommendations aimed at improving the catering management process at schools. The recommendations developed were supported by measurable parameters, which must be monitored by school staff. The capability-driven development approach was applied to model the recommendation system. A plate waste predictive module and a large language model classifier were integrated into the system to support sustainable decision-making. The large language model classifier was trained to filter questions and recommendations. Three training methods were compared: training from scratch and finetuning by using datasets DBPedia and News Category Dataset. As a result, we present the list of recommendations based on the literature review, and the prototype of the knowledge-based recommendation system was developed to audit the school catering management process and promote sustainable school management and decision-making. The recommendation system aims to reduce plate waste due to deficiencies in the implementation of the catering process and to promote responsible food consumption at schools.

The aim of the given scientific paper "Message validation on profanity using LLM” to train LLM to detect profanity in user messages. During the research, the authors created a dataset of 30 phrases (15 good, 15 bad), which were then used to train LLM using AI platform Cohera and evaluated the recognition accuracy. The authors achieved accuracy equal to 70%.

Fruit cultivation is a significant part of the country economics and agriculture. In this paper, aiming to improve apple cultivation, we trained a neural network for apple detection using the YOLOv5 architecture, utilizing the dataset from the lzp-2021/1-0134 project. The dataset consisted of a set of apple tree photographs with apple fruits. Dataset contained 40 images with size 640x640 px. YOLOv5m model was trained five times. The best result model achieved mAP@0.5 equal to 0.9 and mAP@0.5:0.95 equal to 0.63. The artificial intelligence opens new possibilities for horticulture saving resources, which can be redirected on other tasks significantly increasing the efficiency of commercial orchards.

The monitoring of Colorado potato beetles in large fields is a complex and time-consuming process that requires accurate data collection, analysis, and interpretation. The use of artificial intelligence (AI) can greatly simplify this complex process by automatically monitoring fields and detecting beetles marking locations of their location. We trained an object detection neural network using YOLOv5 framework using the dataset with 150 images of the Colorado potato beetles photographed in the natural conditions. We trained YOLOv5s model and got the most perfect result that was 0.995 mAP. Using the best trained neural network, we can now put videos and image files in the code and the neural network will run and detect all the Colorado potato beetles in the given files automatically.

The project aims to leverage blockchain technology to address the challenges faced by gardeners and botanists in tracking and managing plant hybridization data. By employing a decentralized ledger system, the project enables users to securely record and verify the genetic makeup, metadata, and ownership of hybrid plants. Through the integration of smart contracts and MetaMask authentication, users can create and claim ownership of plants, ensuring transparency, integrity, and trust in the hybridization process.