Journal of Artificial Intelligence Research & Advances
Volume 5, Issue 1 (2018)
Published
Table of contents
-
A Robust Face Recognition Model based on Convolutional Neural Networks
Abstract: From the past several years, face recognition is one of the challenging issues in computer vision, due to three main technical problems in it. 1) Expression problem: in which same person shows more than one expression. 2) Illumination problem: in which face images of the person are strongly corrupted by lightning and 3) poses problem: in which large portion of the face becomes invisible due to occlusion. To this end …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 1–11 Read article
-
A Survey on Word Sense Disambiguation for Persian language
Abstract: Natural language processing (NLP) is considered as one of the most important subfields of Artificial Intelligence. It’s used to provide an Interface between Computer and Human languages. NLP has found its application on several areas such as Speech processing, Summation of text, Machine translation, Searching, Information grouping, Parts of Speech Tagging (POS), Sentiment analysis and etc. Natural language processing is posing some challenges in speech recognition, ambiguity in preposition attachment, …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 12–17 Read article
-
Weighted Latent Dirichlet Allocation, an Optimized and Generalized Probabilistic Topic Model
Abstract: In this study, we have introduced a new distribution named as Mixture Weighted Dirichlet Distribution (MWDD) which acts as generalization to other distributions viz. Dirichlet Distribution, Length or Size-Biased Dirichlet, Area Biased and Volume Biased Dirichlet distributions. Aim of this research is to introduce and implement MWDD for Probabilistic Topic Modeling of Corpus of textual data as an optimized as well as generalized probabilistic topic model. Various statistical and structural …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 18–31 Read article
-
An Analytical Review on Word Sense Disambiguation
Abstract: Due to high Semantic Ambiguity that is associated with language, Word Sense Disambiguation is an open problem of Natural Language Processing and Ontology in Computer Linguistics. For instance, the word Head can mean the part of body and a person in charge of organization. The task of automatically assigning the most suitable meaning to a polysemous word can be defined as Word Sense Disambiguation. Several approaches have been done, from …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 32–37 Read article
-
A Novel Software Cost Estimation Model Based on the Hybrid of Artificial Neural Network and Firefly Algorithm
Abstract: Software cost estimation is the prediction of software development effort and software development time required to develop a software project. Software cost estimation is a basic and important task for both successful execution of software development life cycle and its management interms of its cost & time. Accurate software cost estimation is considered to be a difficult task as the information about the software project to be developed at the …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 38–45 Read article
-
PSO tuned SVM-SMC Controller for the Trajectory Tracking of a Five Bar Linkage Manipulator System
Abstract: AbstractRemarkable advantages of SVM make it one of the most widespread and successfully used estimating technique of the era especially for the non-linear systems like robotic manipulators. Control system used in trajectory tracking problems of robotic manipulator is of great importance and hence, accuracy in tracking of manipulator is directly associated to the improvement in the quality of the controller engaged. So, to improve the control performance of the SVM …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 46–55 Read article
-
Efficient Classification of Noisy Text
Abstract: Textual content comprises a significant volume of data generated online on a daily basis. The web-generated data often consists of high levels of noise due to a variety of factors. Development of efficient systems for automatic classification of noisy data is a crucial task in text mining. This paper examines a technique for classification of noisy text which is based on multiple feature selection and supervised learning. The main aim …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 56–61 Read article
-
A Novel Strategy for Weight Initialization in Sigmoidal Feed-forward Artificial Neural Networks
Abstract: In this paper, a novel method of weight initialization is proposed. The proposed method of weight initialization distributes the initial weights and thresholds in such a manner that they lie in different regions of the activation function used at the hidden layer. The proposed method is compared with six other popular weight initialization methods on ten function approximation problems using the RPROP (Resilient Back-propagation) and Levenberg-Marquardt algorithms for training. Two …
Published in Journal of Artificial Intelligence Research & Advances · Vol. 5, Issue 1, 2018 · pp. 62–75 Read article