Article Preview
Top1. Introduction
The most vital task when a lost student is found is to recognize the student. It becomes difficult and time consuming venture when we do not have accurate information about the lost student. Hence a soft computing model can be employed for assisting in identifying the lost student in such environment where detailed information of the student is not available. To the best of our knowledge there has not been any reported work on identifying the lost student on the basis of improper information.
The task of identifying a lost student requires categorization and classification of textual information. Text categorization (also known as text classification or topic spotting) is the task of assigning predefined categories to free-text documents based on their content (Sebastiani, 2002). Automated text classification is desirable because it gives freedom from manually organizing text or document, which can be too expensive, or not feasible due to the time constraint. We can employ text categorization using different methods such as statistical and machine learning techniques which includes support vector machine (SVM) (Cortes & Vapnik, 1995), regression analysis, decision trees, k-nearest neighbor classifier, and Bayesian classifiers (Sahami, 1996).
To deal with imprecise and inexact information a new paradigm has been introduced in computational strategies known as “Soft computing” coined by Lotfi Asker Zadeh (Zadeh, 1965). Soft computing is a consortium of methodologies, which includes Fuzzy Logic (FL), Neural Networks (NN) (Carpenter, Grossberg, Markuzon, Reynolds & Rosen, 1992), and Evolutionary Algorithms (e.g., Genetic algorithms (Goldberg, 1989)) as a main constituents and is also confluence with Expert System (ES) (Buchanan & Smith, 1989), and Machine Learning(ML), which provide flexible information processing capabilities to solve real-life problems. The advantages of employing soft computing (Mitra, Pal, & Mitra, 2002) is its capability to tolerate imprecision, uncertainty, and partial truth (Maeda, Ashida, Taniguchi, & Takahashi, 1995) to achieve tractability and robustness on simulating human decision-making behavior with low cost.
Identifying a lost student with inappropriate information is a tiring and time consuming task. In this paper we have proposed a soft computing model for identifying lost student which takes Student Information as input and identify the lost student. The model of the proposed system is depicted in Figure 1.
Figure 1. Schematic diagram of soft computing model for tracking lost student
The soft computing model requires appropriate representation of student information. The characteristics of lost student information like impreciseness, inappropriate data, and absence of various fields make it suitable for representing it as a Symbolic Object (Bock & Diday, 2000). The Symbolic Object is processed further using a technique called similarity measure to identify the lost student. The similarity measure assigns different similarity values to find out the nearness of the lost student to different student object present in the knowledge base. Hence the similarity measure resembles with fuzzy membership function.
Symbolic objects are extensions of classical data types. A symbolic object and its extent can model a concept or an object of the real world. Symbolic objects can be of three different types, assertion object, hoard object, and synthetic object. An assertion object is a conjunction of events pertaining to a given object. An event is a pair which links feature variables and feature values. A hoard object is a collection of one or more assertion objects, whereas a synthetic object is a collection of one or more hoard objects (Diday, 2000).
The soft computing model is tested for identifying lost student using a knowledge base of 300 student object and test is conducted of 200 lost students and overall 94% of accuracy is achieved.