3225 Slices
Medium 9781601323194

Direction of Arrival Estimation Using Particle Swarm Optimization-based SPECC

Hamid R. Arabnia, Ray R. Hashemi, George Jandieri, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF
Medium 9781601323538

SESSION Late Breaking Papers

Hamid R. Arabnia, Fernando G. Tinetti, Mary Yang CSREA Press PDF

Int'l Conf. on Advances in Big Data Analytics | ABDA'16 |

SESSION

LATE BREAKING PAPERS

Chair(s)

TBA

ISBN: 1-60132-427-8, CSREA Press ©

121

122

Int'l Conf. on Advances in Big Data Analytics | ABDA'16 |

ISBN: 1-60132-427-8, CSREA Press ©

Int'l Conf. on Advances in Big Data Analytics | ABDA'16 |

123

IDEAS: An online tool to Identify

Differential Expression of genes for

Applications in genome-wide Studies

William Yang

School of Computer Science

Carnegie Mellon University

5000 Forbe Ave., Pittsburgh, PA, 15213 U.S.A. wyang1@andrew.cmu.edu

Kenji Yoshigoe, Xiaosheng Wang,

Dan Li, Yifan Zhang

MidSouth Bioinformatics Center and Joint

Bioinformatics Program of University of

Arkansas at Little Rock and University of

Arkansas for Medical Sciences, 2801 S. Univ.

Ave, Little Rock, AR 72204 USA

Patrycja Krakowiak

Wenbing Zhao

Arkansas School for Mathematics and Sciences,

See All Chapters
Medium 9781601322470

Association inference processing to extract knowledge sentences for question answering

Hamid R. Arabnia; David de la Fuente; Elena B. Kozerenko; Peter M. LaMonica; Raymond A. Liuzzi; Todd Waskiewicz; George Jandieri; Ashu M. G. Solo; Ivan Nunes da Silva; Fernando G. Tinetti; and Fadi Thabtah (Editors) Mercury Learning and Information PDF

Int'l Conf. Artificial Intelligence | ICAI'13 |

741

Association inference processing to extract knowledge sentences for question answering

1

Hirokazu Watabe, 2Misako Imono, 3Eriko Yoshimura, and 4Seiji Tsuchiya

Dept. of Intelligent Information Engineering and Science,

Doshisha University, Kyotanabe, Kyoto, 610-0394, Japan

1 hwatabe@mail.doshisha.ac.jp, 2 etl1701@mail4.doshisha.ac.jp, 3 sk109716@mail.doshisha.ac.jp, stsuchiy@mail.doshisha.ac.jp

Abstract - Much of everyday conversation consists of answering questions, with the respondent commonly exhibiting an innate ability to extract appropriate answers from a vast volume of mentally stored knowledge. In computer-based question-answering systems, knowledge extraction has generally involved searching knowledge representations organized by predicate logic or production rules (if-then rules), to find objects of inference. As it is based on symbol processing, however, it is inherently dependent on word notation. In the present paper, we propose a flexible knowledge extraction method in which words having the same meaning but different notations can become objects of inference through incorporation of an association system[1,2] which comprises a Concept-Base that utilizes a method for calculating the degree of association.

See All Chapters
Medium 9781601323217

An Empirical Approach to Identifying the Friendships on Social Network Users

Hamid R. Arabnia, Leonidas Deligiannidis, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Internet Computing and Big Data | ICOMP'14 |

147

An� Empirical� Approach� to� Identifying� the� Friendships� on� Social�Network� Users

Kuan-Hsi Chen1 , Yuh-Jyh Hu1 , Tyne Liang1 , and Kiwing To2

1 Department of Computer Science, National Chiao Tung University, Hsinchu, Taiwan

2 Institute of Physics, Academia Sinica, Taipei, Taiwan

Abstract— Effective friendship identification among users is demanded for most social network services. Nevertheless, such task on a sparse social graph is challenging for network service researchers. Other than the friend database used in most previous researches, the presented study concerns user’s textual interactions corpora, containing user’s posted messages and received responses. From them, multiple types of interaction features are extracted, including user’s sentiment, content topics, response and link information.

The proposed identification is built on a stacking learning mechanism and is verified using real social network data.

Compared to two other supervised learning models, the proposed approach achieves the highest F1-score and the fastest identification speed. Moreover, the experimental results show that the addressed interaction features indeed play positive roles for friendship identification on social network users.

See All Chapters
Medium 9781601323248

Session - Iimage Enhancement Methods, Noise Reduction Algorithms, Image Quality Assessment and Related Technologies

Hamid R. Arabnia, Leonidas Deligiannidis Joan Lu, Fernando G. Tinetti, Jane You, George Jandieri, Gerald Schaefer, and Ashu M.G. Solo CSREA Press PDF

See All Slices