Numerous studies in Mathematics Education have shown that among the causes of disciplinary learning difficulties are the acquisition and understanding of its specialized language.Data-Driven Learning (DDL) is a didactic approach that treats language as data and sees students as researchers doing guided discovery activities. The exploration of corpora can effectively support reflection on the specialized languages of Mathematics. What data should be used? Students daily use the most recent Large Language Models (LLM) and Google, which can be used for linguistic investigations. However, it must be remembered that there is no total control over the data on which searches are carried out, the results, and the type of language they use. Control over the data is important, especially when teachers want to use these tools to design and deliver didactic activities. This paper presents a recent DDL research activity with 80 secondary school students on the specialized language of Mathematics. The students conducted linguistic investigations on a specially designed corpus and carried out corpus-based activities with automatic formative assessment within a Digital Learning Environment. The results show that the students appreciated the proposed activities. They develop linguistic and mathematical skills and become more aware of the importance of the language they use. Students developed digital skills in browsing, searching, and filtering data, as well as in evaluating data, information, and digital content. LLM, such as ChatGPT, could not be used for the same type of activity, but with appropriate design, they can be used as a starting point for investigation and linguistic reflection. In the future, given the notable diffusion of these AI tools, it is essential to train teachers and students on their strengths and weaknesses and how they influence teaching and learning.

Teaching the Specialized Language of Mathematics with a Data-Driven Approach: What Data Do We Use?

Fissore, Cecilia
;
Floris, Francesco
;
Conte, Marina Marchisio
;
Sacchet, Matteo
In corso di stampa

Abstract

Numerous studies in Mathematics Education have shown that among the causes of disciplinary learning difficulties are the acquisition and understanding of its specialized language.Data-Driven Learning (DDL) is a didactic approach that treats language as data and sees students as researchers doing guided discovery activities. The exploration of corpora can effectively support reflection on the specialized languages of Mathematics. What data should be used? Students daily use the most recent Large Language Models (LLM) and Google, which can be used for linguistic investigations. However, it must be remembered that there is no total control over the data on which searches are carried out, the results, and the type of language they use. Control over the data is important, especially when teachers want to use these tools to design and deliver didactic activities. This paper presents a recent DDL research activity with 80 secondary school students on the specialized language of Mathematics. The students conducted linguistic investigations on a specially designed corpus and carried out corpus-based activities with automatic formative assessment within a Digital Learning Environment. The results show that the students appreciated the proposed activities. They develop linguistic and mathematical skills and become more aware of the importance of the language they use. Students developed digital skills in browsing, searching, and filtering data, as well as in evaluating data, information, and digital content. LLM, such as ChatGPT, could not be used for the same type of activity, but with appropriate design, they can be used as a starting point for investigation and linguistic reflection. In the future, given the notable diffusion of these AI tools, it is essential to train teachers and students on their strengths and weaknesses and how they influence teaching and learning.
In corso di stampa
Bridging the Gap Between AI and Reality (AISoLA 2023)
Crete, Greece
23-28 October 2023
Bridging the Gap Between AI and Reality. AISoLA 2023. Lecture Notes in Computer Science
Bernhard Steffen
14129
48
64
9783031737404
9783031737411
Data-Driven Learning, Large Language Model, Language for Specific Purposes, Mathematics Education
Fissore, Cecilia; Floris, Francesco; Conte, Marina Marchisio; Sacchet, Matteo
File in questo prodotto:
File Dimensione Formato  
Teaching_the_Specialized_Language_of_Mathematics_w.pdf

Accesso aperto

Tipo di file: PDF EDITORIALE
Dimensione 2.78 MB
Formato Adobe PDF
2.78 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/2031332
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact