Bogdan Patrut

B R A I N

Broad Research in Artificial Intelligence and Neuroscience

Volume 8, Issue 4 December, 2017

EduSoft

www.brain.edusoft.ro

1

Nitin Balaji Bilgi , Gurusidheshwar M. Wali , Akashta Mense , Priyanka Takkekar , Bhagyashri Yaddi , Sudarshan Patil

Akashta Mense Priyanka Takkekar Bhagyashri Yaddi Sudarshan Patil Students of Department of Computer Science & Engineering, MMEC, Belgaum-591113, India

Symptomatic Decision Support System for Neurological Disorders is a web-based decision support system for diagnosis of neurologic disorders. This system uses Web based decision support system by encoding rules of the neurology domain and also by developing a framework to learn from the cases of the patients. This knowledge encoding is essentially the implementation of two artificial reasoning techniques called Case-Based Reasoning and Rule-Based Reasoning. The system will collect rules of the neurology domain expert and also the case history of the patients. The system will use the rules and cases to achieve the objective of assisting the decision-making process for the domain experts. Decision Support System will give the patient a symptomatic diagnostic conclusion. The system will provide case study of patients and also will display results in form of graphs.

2

Andrei Scutelnicu , Mihaela Onofrei , Anca Diana Bibiri , Mircea Hulea

Automatic speech segmentation has many applications in speech processing and phonetics, e.g., in automatic speech recognition and automatic annotation of speech corpora. In both processes of training and evaluation of speech recognition systems large aligned speech-to-text corpora are needed. Once aligned, identification of phonemes could be based on samples that are picked-up inbetween phonemes' boundaries. Because manual segmentation is costly and extremely time consuming, automatic methods of alignment are searched for. In this paper, we propose a simple, yet efficient, method for speech to text recognition based on a machine learning approach, using a Romanian speech corpus.

3

Ali Taghinezhad , Mehdi Dastpak , Hadi Sameni , Maryam Jamalzadeh , Roya Azadikhah , Mahboobeh Azadikhah , Mohsen Ahkami

The purpose of this study was to investigate empirically the application of one of the cornerstones of chaos theory, that is, sensitivity to initial conditions, to language assessment. In doing so, 20 Iranian students at masters' level participated in this study. They were majoring in nanophysics in Islamic Azad University of Bushehr. They were sophomores aged 23 to 30. In order to ensure homogeneity, they were administered a test of reading comprehension called PET. Then, another test of reading comprehension was administered to them based on "English for the Students of Physics". Having taken that test, the students took the same test after two weeks in different versions. The versions were created by changing the order of items. The analysis of the data showed that 75.83% of students performed better in the second version of the test. This showed the effect of changing the order of items on the students' performance which is consistent with the principle of sensitivity to initial conditions in chaos theory. The implications of the findings will be discussed at the end of the study.

4

Laszlo Csiki , Andrei Gabor , Marius Popescu , Antoanela Naaji

Modern universities across the world rely increasingly on information and communication technologies, both in their educational and research processes, and the management of academic databases. The android app "Virtual University" is a virtual platform designed especially to be used by students and is aimed at accessing, in real time, information connected with the subjects being studied (grades, announcements, notification), financial situation, as well as connecting with the secretariat of the university. The application allows each student to check their financial situation and to pay their educational fees and dues online. The app is meant as an expansion of existing functionalities within universities, and not as a replacement for them. Using the application may lead to an improvement in the quality of life, particularly for people with disabilities.

5

Mohamed Soltane , Lotfi Messikh , Abdelhalim Zaoui

As the information age matures, a biometric identification technology will be at the heart of computer interaction with humans and the biosphere in which they reside. Hence, the reliable information security mechanisms are needed to combat the rising magnitude of identity theft. While cryptography is a powerful tool to achieve information security, one of the main challenges in cryptosystems is to maintain the secrecy of the cryptographic keys. Template protection techniques prevent stored reference data from revealing private biometric information and enhance the security of biometric systems against attacks such as identity theft and cross matching. A critical issue in biometric systems is to protect the template of a user which is typically stored in a database or a smart card. The fuzzy vault construct is a challenging biometric cryptosystem that secures both the secret key and the biometric template by binding them within a cryptographic framework. The helper data itself do not leak any information about the biometric template, yet contain sufficient information to align the template and query biometric accurately. This paper reviews the state of the art biometrics Cryptosystems from the Point of Challenging Designs Strategies.

6

Fateme Maleki , Majid Yousefikhoshbakht , Amin Rahati

This paper presents a hybrid self-adaptive global best harmony search algorithm (HSGHSA) to improve the performance of harmony search algorithm (HSA) for solving vehicle routing problems with time windows (VRPTW). To explore the search space more efficiently, the proposed HSGHSA couples an improved variant of HSA called global best harmony search algorithm with a self-adaptive mechanism for tuning its control parameters. Moreover, the HSGHSA adopts six local search (LS) neighborhood structures to enhance its exploitation capability. The effectiveness of HSGHSA is evaluated against Solomon's VRPTW benchmark and its performance is compared with HSA and several state-of-the-art algorithms. The obtained results confirm that the HSGHSA produces very competitive results compared to the other algorithms.

7

Angel Garrido

From ancient times, the history of human beings has developed through a succession of steps and sometimes jumps, until reaching at the relative sophistication of the modern brain and culture. Researchers are attempting to create systems that mimic human thinking, understand speech, or beat the best human chess player. Understanding the mechanisms of intelligence, and creating intelligent artefacts are the twin goals of Artificial Intelligence (AI). Great mathematical minds have played a key role in AI in recent years; to name only a few, Janos Neumann (also known as John von Neumann), Konrad Zuse, Norbert Wiener, Claude E. Shannon, Alan M. Turing, Grigore Moisil, Lofti A. Zadeh, Ronald R. Yager, Michio Sugeno, Solomon Marcus, or Lászlo A. Barabási. Introducing the study of AI is not merely useful because of its capability to solve difficult problems, but also because of its Mathematical nature. It prepares us to understand the current world, enabling us to act on the challenges of the future.

8

Iuliana Hunea , Simona Irina Damian , Sofia David , Madalina Maria Diac , Diana Bulgaru Iliescu , Manuela Ciocoiu

Current work reviews the concept, pathological mechanism and the process of diagnosing diffuse axonal lesions (DAI). The pathological mechanism underlying DAI is complicated, including axotomy with the appearance of retraction bulbs, interrupted protein transport along axonal neurophilaments, massive influx of calcium ions and calpain-mediated hydrolysis, axonal cytoskeletal cell degradation, beta-amyloid precursor protein accumulation and changes in glial cells. The transition from primary axotomy to secondary axotomy causes a very complex, calciumdependent biochemical cascade.

9

Huseyin Bicen , Mobina Beheshti

Nowadays, rapid changes in technology have significant influence in learners' educational life. The technological devices of information and communication are developed to deliver valuable knowledge quickly, regardless of the place and time, novel media demonstration formats emerged. Infographics are examples of this format, which use graphic visual pictures to show the information, knowledge or data effectively. Infographics are used in instruction, particularly in instructional design which is more challenging to design an education. Hence, teaching by infographics helps students to interpret visual knowledge and provide a broader and extensive body of learning and grasp in education. The main objective of this research is to investigate the students' perceptions about using infographics in education. The research is designed as a quantitative study. The total number of undergraduate students participating in this research was 163. Data were accumulated with close-ended questions during this study. The gathered data were analysed via the descriptive analysis approach.

10

Dattaprasad Torse , Veena Desai , Rajashri Khanai

At present, manual observation of the electroencephalogram (EEG) signals is the prime method for diagnosis of epileptic seizure disorders. The method is a time consuming and error prone as it involves errors due to fatigue in continuous monitoring of nonlinear and nonstationary EEG signals. Out of approximate 1% of the world's epilepsy patients more than 25% cannot be treated correctly due to erroneous diagnosis. The automated seizure detection system can prove efficient by making the process reliable and faster. This paper reviews multi-domain feature extraction and machine learning classification techniques used in automated seizure detection systems. To analyse subtle variations in EEG, signal decomposition algorithms have been used in time, frequency, joint time-frequency, and nonlinear domain. The statistical and entropy parameters are the key features to discern normal from the seizure EEG signals. Machine learning plays a critical role in extracting meaningful information out of the extracted features. The paper also evaluates the performance of Multilayer Perceptron Neural Network, naïve Bayes, Least Square Support Vector Machine, k nearest neighbour, and random forest classifiers using sensitivity, specificity and accuracy metrics. A seizure detection technique is developed by decomposing the EEG signals by means of Tunable-Q Wavelet Transform (TQWT). To quantify the complexity of the individual multivariate sub-bands of the biomedical signals TQWT proves effective with varied values of Q factor suitable for analyzing signals with oscillatory and non-oscillatory nature. The highest accuracy of 97.3% is obtained using random forest classifier for the combination of spectral, Shannon and Kraskov entropy features. The paper compares the performance of feature extraction and classification techniques for the implemented system. The comparison explores possibility of hardware implementation of real time seizure detection scheme.