The paper introduces Verificator, our learning programming interface aimed for learning programming in C++ at the university beginners' level. In teaching programming some specific problems concerning the teaching itself as well as the organization of the teaching process need to be considered. One of the biggest problems is that students tend to adopt certain bad programming habits in their attempt to more easily deal with their examinations, such as trying to write programs without any syntax and logical checking. It is very hard to help them correct those errors once they are deeply rooted. Our students' web questionnaire and its results show that the majority of problems in learning programming among our students arise from the gap between the understanding of programming language syntax and problem-solving algorithms. Verificator prevents students from making a lot of errors they are likely to make in learning programming and helps them to learn programming language syntax and adopt good programming habits.

Programming education traditionally requires extensive manual assessment of student assignments, which is both time-consuming and resource-intensive for instructors. Recent advances in large language models (LLMs) open opportunities for automating this process and providing timely feedback. This paper investigates the application of artificial intelligence (AI) tools for preliminary assessment of undergraduate programming assignments. A multi-phase experimental study was conducted across three computer science courses: Introduction to Programming, Programming 2, and Advanced Programming Concepts. A total of 315 Python assignments were collected from the Moodle learning management system, with 100 randomly selected submissions analyzed in detail. AI evaluation was performed using ChatGPT-4 (GPT-4-turbo), Claude 3, and Gemini 1.5 Pro models, employing structured prompts aligned with a predefined rubric that assessed functionality, code structure, documentation, and efficiency. Quantitative results demonstrate high correlation between AI-generated scores and instructor evaluations, with ChatGPT-4 achieving the highest consistency (Pearson coefficient 0.91) and the lowest average absolute deviation (0.68 points). Qualitative analysis highlights AI’s ability to provide structured, actionable feedback, though variability across models was observed. The study identifies benefits such as faster evaluation and enhanced feedback quality, alongside challenges including model limitations, potential biases, and the need for human oversight. Recommendations emphasize hybrid evaluation approaches combining AI automation with instructor supervision, ethical guidelines, and integration of AI tools into learning management systems. The findings indicate that AI-assisted grading can improve efficiency and pedagogical outcomes while maintaining academic integrity.

The paper presents a case study on using artificial intelligence (AI) for preliminary grading of student programming assignments. By integrating our previously introduced learning programming interface Verificator with the Gemini 2.5 large language model via Google AI Studio, C++ student submissions were evaluated automatically and compared with teacher-assigned grades. The results showed moderate to high correlation, although the AI was stricter. The study demonstrates that AI tools can improve grading speed and consistency while highlighting the need for human oversight due to limitations in interpreting non-standard solutions. It also emphasizes ethical considerations such as transparency, bias, and data privacy in educational AI use. A hybrid grading model combining AI efficiency and human judgment is recommended.

Publisher: Vilnius University Institute of Data Science and Digital Technologies, Journal: Informatica, Title: Dynamic Generation of Web Services for Data Retrieval Using Ontology, Authors: Ivan Magdalenic, Danijel Radosevic, Zoran Skocir , Semantic Web is envisioned as semantic description of data and services enabling unambiguous computerized interpretation. Thanks to semantic description, computers can perform demanding tasks such as automation of discovery and access to heterogeneous data sources. Although this is possible with the existing technologies, combination of web services technology, ontologies and generative programming methods makes this simpler and more efficient. This paper presents the model for dynamic generation of web services for data retrieval from heterogeneous data sources using ontologies. Emphasis is on dynamic generation of web services customized to a particular user based on the request defined by ontology. The paper also describes a prototype of the model implementation. Some advantages of our approach over other approaches are also provided.

We revisit the old but formally still unresolved debate on the time efficiency of accessing the elements of 1D arrays via indices versus accessing them via pointers. To analyze that, we have programmed benchmarks of minimal complexity in the C++ language and inspected the machine code of their compilation in the x86 assembly language. Before exploring the performance, we briefly compared a few methods used for the execution time measurements. The results on the Wintel platform show no significant advantage in using pointers over indices except for some benchmarks and array (data) types. In other cases, the exact opposite may be true. The cause of this inconsistency lies in the compilation of the source code into the rather nonorthogonal x86 instruction set. Furthermore, the execution speed does not clearly relate to the instruction length. The parallel aim of this work is to provide a ground for further analysis and measurements of this kind using different compilers, languages, and computer platforms.