Sari Jabbarin

Generating Use Case Diagrams from Arabic User Requirements in a Semi-automated Approach Using a Natural Language Processing Tool

Automated software engineering has attracted a large amount of research efforts. The use of object-oriented methodologies for software systems development has made it necessary to develop approaches that automate the construction of different UML models in a semi-automated approach from textual user requirements. UML use case models represent an essential artifact that provides a view of the system under analysis or development. The development of such use case models is very crucial in an object-oriented development methodology. The main principles used in obtaining these models are described. A natural language processing tool is used to parse different statements of the user requirements written in Arabic language to obtain lists of nouns, noun phrases, verbs, verb phrases, etc. that aid in finding  potential  actors and use cases. A set of heuristics/steps that represent our approach for constructing a use case model is presented. Finally, the proposed approach is validated using an experiment involving a group of graduate students who are familiar with use case modeling.

 

Muna

Student Modeling in Adaptive Educational Chat Room

In recent years, universities widely started to use E-learning in the educational process. Some of these systems are hybrid while others are pure E-learning. Through E-learning the student can get the teaching materials without the intervention of the teacher, which enables him to study anytime and from anywhere.

Virtual Classrooms are among the best tools used in E-learning. Moodle is an example of such virtual classroom.

The main problem in using virtual classrooms is the diversity in the student's attributes who attend the virtual classroom. The virtual classroom usually presents the LOs and TOs in the same way for all the students.

Adaptive Virtual Classroom Model (AVCM) discussed in the thesis is a model which introduced a solution for the previous problem. AVCM enables the student to receive adaptive presentations, and then progress to adaptive tests, taking into account the student’s personal characteristics. After the completion of the concept presentations in the virtual classroom, and the adaptive test as well, the student should attend the chat session to discuss the same concept with one of his peers. This discussion assumes to have as an output an assessment of the student, as one of the inputs used in the evaluation process of the student's final grade in the virtual classroom.

The chat room proposed by AVCM enables one to one discussion about one and only one concept. It is considered as one main component in the educational process which should be monitored and evaluated using computers science techniques specially database management systems, programming languages and artificial intelligence abilities.

This research aimed to find a viable way to model the student's behavior during the chat session, which was commissioned by the teacher to spend with one of his fellow students. The session is usually done after the completion of the presentation of a particular concept through the virtual classroom. Student modeling in the chat session means to have two main outputs namely. 1) A final grade for the student as an evaluation score extracted from the student’s usage of time in the session. 2) The student’s status during the current session which will help his next peer in the next session to know more about him and conduct his discussion based on this status.

Computer science and informatics play a vital role in this research. Computer science will help us in chat time and text analysis, statistical and regression analysis, in addition to building the algorithms, database construction, storage and retrieval media. Artificial intelligence techniques namely the rule base systems should be used to construct the adaptive user interface basing on the results obtained by student modeling. The model should include knowledge base and a large number of rules for reasoning process. At the end of this research, an algorithm will be constructed for the whole modeling and reasoning process.

Previous studies, experiments and experts were used to extract the evaluation criteria. The study was able to get a methodology applicable in the assessment process. This methodology is based on three models, namely: Time model, Text model, and Peer model. All these models parameters can be obtained through mathematical equations created through statistical procedures. Three mathematical equations contain a set of variables together constitute the specific ratios within and final evaluation of the student in a chat room.

       By this evaluation process, we achieved two main objectives: the first one is the student’s assessment during a chat session, and the second goal is to have a formalized chat room, making the students dealing more seriously with the discussion. Taking into consideration that everything that comes from the student during the session; whether the extent of exploitation for the time of the session effectively, or the words and expressions used by him and his colleague. All these actions will affect the evaluation process.