Search
Search Results
-
A Nim like game and a machine that plays it: a learning situation at the interface of mathematics and computer science
317-326Views:96The purpose of this work is to take a didactic look at a learning situation located at the interface between mathematics and computer science. This situation offers a first approach to the concept of artificial intelligence through the study of a reinforcement learning device. The learning situation, inspired by the Computer Science Unplugged approach, is based on a combinatorial game, along with a device that learns how to play this game. We studied the learning potential when the human players face the machine. After an a priori analysis using the Theory of Didactic Situations (TDS), we conducted a pre-experiment in order to strengthen our hypotheses. In this article, we will focus on the analysis of the didactic variables, the values we have chosen for these variables and their effects on students’ strategies.
Subject Classification: 97D99, 97K99, 97P80
-
Proof step analysis for proof tutoring - a learning approach to granularity
325-343Views:6We present a proof step diagnosis module based on the mathematical assistant system Ωmega. The task of this module is to evaluate proof steps as typically uttered by students in tutoring sessions on mathematical proofs. In particular, we categorise the step size of proof steps performed by the student, in order to recognise if they are appropriate with respect to the student model. We propose an approach which builds on reconstructions of the proof in question via automated proof search using a cognitively motivated proof calculus. Our approach employs learning techniques and incorporates a student model, and our diagnosis module can be adjusted to different domains and users. We present a first evaluation based on empirical data. -
An improvement of the classification algorithm results
131-142Views:6One of the most important aspects of the precision of a classification is the suitable selection of a classification algorithm and a training set for a given task. Basic principles of machine learning can be used for this selection [3]. In this paper, we have focused on improving the precision of classification algorithms results. Two kinds of approaches are known: Boosting and Bagging. This paper describes the use of the first method – boosting [6] which aims at algorithms generating decision trees. A modification of the AdaBoost algorithm was implemented. Another similar method called Bagging [1] is mentioned. Results of performance tests focused on the use of the boosting method on binary decision trees are presented. The minimum number of decision trees, which enables improvement of the classification performed by a base machine learning algorithm, was found. The tests were carried out using the Reuters 21578 collection of documents and documents from an internet portal of TV Markíza. -
Teaching Java programming using case studies
245-256Views:2The paper deals with the technical background and the pedagogical issues of a specific implementation for the collection, assessment and archiving of the students' assignments written in Java. The implemented system automatically applies object-oriented metrics on the collected works in order to measure the characteristic features of the assignments. Tutors use these results for the detection of plagiarisms and for the selection of outstanding works. The paper interprets the measured values within a real Java course held in the 3rd term of the Informatics bachelor study programme at the technical university. Students have several case studies devoted to the simulation of the ATM (Automatic Teller Machine) at disposal. We conclude that the access to the analyzed pool of case studies, blended with the Sun Learning Connection license from the Sun Microsystems, Inc., is an effective way of teaching programming in Java.