Dewi, Aulia Rahma
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

PROBLEMATIKA PENALARAN CHATGPT DALAM MENYELESAIKAN SOAL KOMBINATORIKA [THE PROBLEMATIC REASONING OF CHATGPT IN SOLVING COMBINATORICS PROBLEMS] Rofiki, Imam; Dewi, Aulia Rahma
JOHME: Journal of Holistic Mathematics Education Vol. 8 No. 2 (2024): DECEMBER
Publisher : Universitas Pelita Harapan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19166/johme.v8i2.8600

Abstract

The development of the digital era in the 21st century has brought significant changes in various aspects of human life, especially education. Technologies such as artificial intelligence (AI) have been applied in various fields, including education, to improve the efficiency and effectiveness of learning. ChatGPT, one of the text-based AI models, is used to answer various user questions. This study aims to analyze ChatGPT's problematic reasoning in solving combinatorics problems. The research method used is qualitative with a case study design. Data were collected through a test and interview (questions and answers) to ChatGPT version 3.5. The results showed that ChatGPT's problems were errors in calculating the number and position of letters, as well as inconsistency in the answers given. Although ChatGPT can provide explanations of basic concepts in combinatorics, ChatGPT has limitations in in-depth reasoning and lacks precise calculation accuracy. In fact, ChatGPT was unable to answer the questions clearly. This shows that ChatGPT lacks reasoning towards solving combinatorics problems. Users need to evaluate and verify the responses/answers given by ChatGPT.BAHASA INDONESIA ABSTRACT: Perkembangan era digital pada abad ke-21 telah membawa perubahan signifikan dalam berbagai aspek kehidupan manusia, khususnya bidang pendidikan. Teknologi seperti kecerdasan buatan/artificial intelligence (AI) telah diterapkan dalam berbagai bidang, termasuk pendidikan, untuk meningkatkan efisiensi dan efektivitas pembelajaran. ChatGPT, sebagai salah satu model AI berbasis teks, digunakan untuk menjawab berbagai pertanyaan pengguna. Penelitian ini bertujuan untuk menganalisis problematika penalaran ChatGPT dalam menyelesaikan soal-soal kombinatorika. Metode penelitian yang digunakan adalah kualitatif dengan desain studi kasus. Data dikumpulkan melalui tes dan wawancara (tanya jawab) terhadap ChatGPT versi 3.5. Hasil penelitian menunjukkan bahwa problematika ChatGPT adalah kesalahan dalam perhitungan jumlah dan posisi huruf, serta ketidakonsistenan dalam jawaban yang diberikan. Meskipun, ChatGPT mampu memberikan penjelasan konsep dasar dalam kombinatorika, tetapi ChatGPT memiliki keterbatasan dalam penalaran mendalam dan tidak memiliki akurasi perhitungan yang tepat. Bahkan ChatGPT tidak dapat menjawab soal yang diberikan secara jelas. Hal ini menunjukkan bahwa ChatGPT memiliki penalaran yang kurang terhadap penyelesaian soal kombinatorika. Pengguna perlu untuk melakukan evaluasi dan verifikasi respons/jawaban yang diberikan oleh ChatGPT.
Analysis of junior high school students' critical thinking skills in solving HOTS questions viewed from learning style Dewi, Aulia Rahma; Sa'dijah, Cholis; Rofiki, Imam; Mustakim, Siti Salina
Jurnal Kajian Pembelajaran Matematika Vol 8, No 2 (2024): JURNAL KAJIAN PEMBELAJARAN MATEMATIKA
Publisher : UNIVERSITAS NEGERI MALANG

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17977/um076v8i22024p85-105

Abstract

Currently junior high school students' critical thinking abilities are relatively low. One way to optimize students' critical thinking skills is through HOTS (High Order Thinking Skills) questions. Students' critical thinking abilities in solving HOTS questions vary due to differences in learning styles. This research analyzes junior high school students' critical thinking abilities in solving HOTS questions regarding learning style. The research method used is descriptive qualitative. Data collection techniques were carried out by distributing questionnaires, tests, and interviews. The interactive analysis model uses data collection, reduction, display, and conclusion drawing/verification. The research results show that students with a visual learning style tend to write answers completely and systematically, can identify problems in a structured manner, and have high accuracy, so they often recheck students' answers. Students with an auditory learning style have difficulty writing complete and structured answers but have good storytelling skills and often explain answers verbally even though they do not write them in detail. Kinesthetic students tend to write answers in sketch form, only recording a few steps and do not double-check answers because they cannot sit still for too long