Influence of Prompts Structure on the Perception and Enhancement of Learning through LLMs in Online Educational Contexts DOI Creative Commons

Silvia Rodriguez-Donaire

IntechOpen eBooks, Год журнала: 2024, Номер unknown

Опубликована: Окт. 15, 2024

This research examines how the structure of prompts impacts perceived depth and accuracy responses generated by generative Large Language Models (LLMs) in educational settings. It specifically investigates prompt design influences students’ learning experiences. The study involved an experiment with 183 students enrolled a mandatory Business Administration course at Universitat Oberta de Catalunya (UOC). Data from were analyzed using both qualitative quantitative methods. results show that well-structured significantly improve perception GenAI-generated responses, leading to more effective process. underscores crucial role maximizing effectiveness GenAI. findings suggest thoughtful can enhance outcomes, although study’s limited sample size context-specific nature may restrict generalizability results. contributes field highlighting importance harnessing GenAI tools for improvement.

Язык: Английский

A Scoping Survey of ChatGPT in Mathematics Education DOI Creative Commons
Birgit Pepin, Nils Buchholtz, Ulises Salinas-Hernández

и другие.

Digital Experiences in Mathematics Education, Год журнала: 2025, Номер unknown

Опубликована: Фев. 17, 2025

Язык: Английский

Процитировано

1

Can Generative AI and ChatGPT Break Human Supremacy in Mathematics and Reshape Competence in Cognitive-Demanding Problem-Solving Tasks? DOI Creative Commons
Deniz Kaya, Selim Yavuz

Journal of Intelligence, Год журнала: 2025, Номер 13(4), С. 43 - 43

Опубликована: Апрель 2, 2025

This study investigates the potential of generative artificial intelligence tools in addressing cognitive challenges encountered by humans during problem-solving. The performance ChatGPT-4o and GPT-4 models NAEP mathematics assessments was evaluated, particularly relation to demands placed on students. Sixty assessment tasks, coded field experts, were analyzed within a framework complexity. provided responses each question, which then evaluated using NAEP’s scoring criteria. study’s dataset average scores students who answered correctly item-wise response percentages. results indicated that outperformed most individual items assessment. Furthermore, as demand increased, higher required answer questions correctly. trend observed across 4th, 8th, 12th grades, though did not demonstrate statistically significant sensitivity increased at 12th-grade level.

Язык: Английский

Процитировано

0

Role of Mathematics Teachers in Learner’s Diversity Using AI Tools DOI Creative Commons
Wo Sang Wilson Cheng

Опубликована: Фев. 26, 2025

Язык: Английский

Процитировано

0

A scoping survey of ChatGPT in mathematics education DOI
Birgit Pepin, Nils Buchholtz,

Ulises Salinas-Fernandez

и другие.

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Сен. 26, 2024

Abstract This initial article of the Special Issue on Chat GPT in mathematics education is two parts: (1) a report scoping review study that provides background to articles Issue; and (2) editorial affords glance at seven Issue.

Язык: Английский

Процитировано

1

Using large language models to support pre-service teachers mathematical reasoning—an exploratory study on ChatGPT as an instrument for creating mathematical proofs in geometry DOI Creative Commons
Frederik Dilling, Marc Herrmann

Frontiers in Artificial Intelligence, Год журнала: 2024, Номер 7

Опубликована: Окт. 23, 2024

In this exploratory study, the potential of large language models (LLMs), specifically ChatGPT to support pre-service primary education mathematics teachers in constructing mathematical proofs geometry is investigated. Utilizing theoretical framework instrumental genesis, prior experiences students with LLMs, their beliefs about operating principle and interactions chatbot are analyzed. Using qualitative content analysis, inductive categories for these aspects formed. Results indicate that had limited LLMs used them predominantly applications not specific. Regarding beliefs, most show only superficial knowledge technology misconceptions common. The analysis showed multiple types parts mathematics-specific prompts patterns on three different levels from single whole chat interactions.

Язык: Английский

Процитировано

1

Communicative AI Agents in Mathematical Task Design: A Qualitative Study of GPT Network Acting as a Multi-professional Team DOI Creative Commons
Sebastian Schorcht, Franziska Peters,

Julian Kriegel

и другие.

Digital Experiences in Mathematics Education, Год журнала: 2024, Номер unknown

Опубликована: Окт. 24, 2024

Abstract This study explores the application of communicative AI agents, specifically a network customized generative pretrained transformer in designing mathematical tasks. It focuses on how these functioning as multi-professional team, can perform task design (concerning collection activities and not curriculum materials/textbooks) through collaborative context-aware communication. Concentrating four perspectives—mathematical depth, language sensitivity, natural differentiation, competence orientation—four different agents were instructed to evaluate modify six tasks based individual research knowledge bases. In consensus-seeking process, connected via chat chain, prompting multiple iterations The output (six AI-modified tasks) was then evaluated by in-service teachers human experts making them choose blindly between original analyzing additional comments their decisions qualitative content analysis. Furthermore, rated multidimensional Likert scale. results indicate that for tasks, achieving balance substantial text generation precise formulation is crucial always found GPT output. At same time, combination able enrich with potential solution approaches specific calls action.

Язык: Английский

Процитировано

1

Influence of Prompts Structure on the Perception and Enhancement of Learning through LLMs in Online Educational Contexts DOI Creative Commons

Silvia Rodriguez-Donaire

IntechOpen eBooks, Год журнала: 2024, Номер unknown

Опубликована: Окт. 15, 2024

This research examines how the structure of prompts impacts perceived depth and accuracy responses generated by generative Large Language Models (LLMs) in educational settings. It specifically investigates prompt design influences students’ learning experiences. The study involved an experiment with 183 students enrolled a mandatory Business Administration course at Universitat Oberta de Catalunya (UOC). Data from were analyzed using both qualitative quantitative methods. results show that well-structured significantly improve perception GenAI-generated responses, leading to more effective process. underscores crucial role maximizing effectiveness GenAI. findings suggest thoughtful can enhance outcomes, although study’s limited sample size context-specific nature may restrict generalizability results. contributes field highlighting importance harnessing GenAI tools for improvement.

Язык: Английский

Процитировано

0