1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 60 61 62 63 64 65 IEEE INDICON 2015 1570203949 Fuzzy Viva Assessment Process through Perceptual Computing Manvi Madan Department of Computer Science G. G. S. I. P. University New Delhi, India e-mail: manvimadan@outlook.com Abstract- Viva assessments are an important part of the curriculum in all the professional and academic courses. Since ages, the processes has been handled manually and hence has been prone to errors and mal practises. Considering the wide usage of the procedure in determining the performance of the individuals and the fuzzy nature of the evaluation process, we have presented an algorithm to calculate the viva assessment scores. In our paper, the system presented calculates the viva assessment scores based on the fuzzy linguistic parameters associated with the assessment. Since a student represents his/ her ideas in the form of "words" in a viva assessment therefore the words associated with the assessment of th� student by the respective authorities provides the viva assessment scores. Keywords- Perceptual Computing (Per-); Viva; Viva Assessment; CW I. I NTRODUCTION The education systems are evolving with the advent of new teclmologies. New milestones were achieved in the development of education support systems with the invention and establishment of online examination systems at various institutes such as University of Pune [1], lawaharlal Nehru University, Hyderabad [2], mobile examination systems for various operating systems [3] [4]. The development of these examination systems has enhanced the eiciency of the assessment of the students due to the implementation of robust algorithms that test the capability of the students in real time [5]. Apart rom the subjective/ objective examinations, the practice of conducting Viva assessments is also widely used in various educational institutions. The assessment process in a viva is ambiguous and leaves grounds for various mal-practices due to heavy reliance on invigilator's perspective. The signiicant cause of imprecise evaluation by an invigilator can be the fact that "words mean different things to different people" [7]. In our paper, we present an algorithm to determine the viva evaluation scores for the students. Since the students represents their ideal knowledge/ view on the subject matter in the form of "words" in a viva assessment, the will be evaluated by processing the linguistic parameters associated with the viva assessment. The traditional viva process includes the student's verbal assessment based on the observation and evaluation made by the invigilator or the invigilation panel. The viva scores given the invigilators have so far been crisp in nature. Though the parameters govening the viva evaluation such Piyush Madan Doon Public School Central Board of Secondary Education New Delhi, India e-mail: madanpiyush5@gmail.com as spontaneity of the student while responding, appropriateness of the answer, clarity in speech, irmness of opinion et. Al are fuzzy in nature. Hence, in our algorithm, the uzziness of these parameters is preserved by taking input in the form of "words" in place of crisp values. The algorithm uses the teclmique of CWW (Computing with Words) to process the linguistic parameters associated with the process of evaluation of viva. Our system is based on the algorithm of perceptual computing [6]. In our algorithm, the input is provided to the encoder of the system in the form of "words". The processing of these linguistic parameters is done by the CWW Engine of the system. The codebook and the Footprints of Uncertainty (FOUs) are plotted for each linguistic parameter by the CWW engine. The decoder provides the viva evaluation score by making required calculations over the processed linguistic variables. The system can be employed at schools and colleges to ensure eficient viva evaluations. The system can also be employed to provide analysis of viva during project presentations in professional courses. Therefore, the system is an optimized solution for eicient viva assessments. In our paper, the results obtained by employing the system for the viva assessment of a batch of 35 students are presented. Rest of the paper is organized as follows: section II focuses on the related work. Section III provides the details of the mathematical preliminaries relevant for the understanding of our system. Section IV provides the details of our algorithm along with the various components of our system. Section V presents the results for the viva assessment of a batch of students and the discussion of the same. In section VI, we conclude our paper and throw some light on the future scope of our work. II. RELATED WOK A number of existing works in the domain of fuzzy logic and educational system has been studies before and during the designing of the system. A literature survey of the existing works has been presented below: In [3], the author has presented an examination system for mobile platform. The system has been designed exclusively for Nokia phones equipped with Symbian OS. The system included majorly all the modules of the examination process along with the facility of generation of question paper automatically by picking up questions rom the question bank. The system is well tested and an advancement for the exam conducting teclmology.