Wyniki wyszukiwania - Biblioteka Nauki

1

A method for learning scenario determination and modification in intelligent tutoring systems

100%

Kozierkiewicz-Hetmańska A. , Nguyen N.

|

tom 21

|

nr 1

69-82

EN

Computers have been employed in education for years. They help to provide educational aids using multimedia forms such as films, pictures, interactive tasks in the learning process, automated testing, etc. In this paper, a concept of an intelligent e-learning system will be proposed. The main purpose of this system is to teach effectively by providing an optimal learning path in each step of the educational process. The determination of a suitable learning path depends on the student's preferences, learning styles, personal features, interests and knowledge state. Therefore, the system has to collect information about the student, which is done during the registration process. A user is classified into a group of students who are similar to him/her. Using information about final successful scenarios of students who belong to the same class as the new student, the system determines an opening learning scenario. The opening learning scenario is the first learning scenario proposed to a student after registering in an intelligent e-learning system. After each lesson, the system tries to evaluate the student's knowledge. If the student has a problem with achieving an assumed score in a test, this means that the opening learning scenario is not adequate for this user. In our concept, for this case an intelligent e-learning system offers a modification of the opening learning scenario using data gathered during the functioning of the system and based on a Bayesian network. In this paper, an algorithm of scenario determination (named ADOLS) and a procedure for modifying the learning scenario AMLS with auxiliary definitions are presented. Preliminary results of an experiment conducted in a prototype of the described system are also described.

2

A method for learning scenario determination and modification in intelligent tutoring systems

75%

Kozierkiewicz-Hetmańska A. , Nguyen N. T.

|

tom Vol. 21, no 1

69-82

EN

Computers have been employed in education for years. They help to provide educational aids using multimedia forms such as films, pictures, interactive tasks in the learning process, automated testing, etc. In this paper, a concept of an intelligent e-learning system will be proposed. The main purpose of this system is to teach effectively by providing an optimal learning path in each step of the educational process. The determination of a suitable learning path depends on the student's preferences, learning styles, personal features, interests and knowledge state. Therefore, the system has to collect information about the student, which is done during the registration process. A user is classified into a group of students who are similar to him/her. Using information about final successful scenarios of students who belong to the same class as the new student, the system determines an opening learning scenario. The opening learning scenario is the first learning scenario proposed to a student after registering in an intelligent e-learning system. After each lesson, the system tries to evaluate the student's knowledge. If the student has a problem with achieving an assumed score in a test, this means that the opening learning scenario is not adequate for this user. In our concept, for this case an intelligent e-learning system offers a modification of the opening learning scenario using data gathered during the functioning of the system and based on a Bayesian network. In this paper, an algorithm of scenario determination (named ADOLS) and a procedure for modifying the learning scenario AMLS with auxiliary definitions are presented. Preliminary results of an experiment conducted in a prototype of the described system are also described.

3

Sztuczna inteligencja (SI) w badaniach lingwistycznych

63%

Sztuk A.

|

nr 25/4

159-174

EN

The main purpose of the paper is both to present and to highlight the wide range of artificial intelligence appliance in linguistic research. I intend to define the so called ‘linguistic intelligence’ in the sense of machine learning, based mainly on artificial neural networks. Linguistic intelligent solutions seem to be not only up-to-date but also very promising in the area of developing and improving any intelligent linguistic tools, such as intelligent tutoring systems that are able to interact with human being, or the voice (speech) recognition systems that are able to receive, interpret (understand) and sometimes even carry out spoken commands. Finally, I intend to present the area of so called ‘terminotics’. The term refers to the meeting point of three interrelated disciplines: terminology, computational linguistics and linguistic engineering. This branch is also assisted by computer tools and new technologies based on artificial intelligence and machine learning. These (tools) are mainly designed for term extraction and corpora development but lately there are also some new possibilities to use these tools to increase the quality of terminology infrastructure as well.