28 Slices
Medium 9781601324078

Late Poster Papers

Editied By Hamid R. Arabnia, David de la Fuente, Roger Dziegiel, Elena B. Kozerenko, Peter M. LaMonica,Raymond A. Liuzzi, Jose A. Olivas, Todd Waskiewicz, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti (Editors) CSREA Press PDF

Int'l Conf. Artificial Intelligence | ICAI'15 |

825

Predicting Correctness of “Google Translate”

Yulia Rossikova, J. Jenny Li, and Patricia Morreale

Computer Science, Kean University

1000 Morris Ave, Union NJ 08340 USA juli@kean.edu

Abstract— This paper presents a new modeless approach for Machine Learning predictions, called Radius of

Neighbors (RN). We applied RN to predict the correctness of Google translator and found it to be an improvement over K-Nearest Neighbors (KNN) in terms of prediction accuracy. Both methods are applicable to situations when a mathematical prediction model does not exist or is unknown. With RN, we will be able to create new applications that rely on the users’ awareness of translation accuracy, e.g. an online instant messager, which allows users to chat in various natural languages.

Keywords – K-Nearest Neighbors (KNN), Machine

Learning, Prediction.

EXTENDED ABSTRACT

Machine learning in recent years has given us many new breakthrough applications such as self-driving cars, phone contact center voice recognision, effective web search and automatic natural language translators. Prediction is a key feature of machine learning and K-Nearest-Neighbor (KNN) is a well known prediction method that doesn't need any known model in advance, and thus is suitable for the situations when the model doesn't exist or is unknown. However the accuracy of KNN hinders his wider usage in prediction.

See All Chapters
Medium 9781601324078

Workshop- Intelligent Linguistic Technologies, Ilintec '15

Editied By Hamid R. Arabnia, David de la Fuente, Roger Dziegiel, Elena B. Kozerenko, Peter M. LaMonica,Raymond A. Liuzzi, Jose A. Olivas, Todd Waskiewicz, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti (Editors) CSREA Press PDF

Int'l Conf. Artificial Intelligence | ICAI'15 |

281

Parametrizing Verb Second Languages and Clitic Second

Languages

Anton Zimmerling 1

1

Institute for Modern Linguistic Research, Sholokhov Moscow State University for the Humanities / Institute of Linguistics, Russian Academy of Sciences, Moscow, Russia

Abstract - In this paper, I discuss verb second (V2) and clitic second languages (CL2) as an object of the parametric typology. V2 systems are a small group of languages sharing a number of parameters constraining the clausal architecture and the finite verb placement. CL2 systems are a large group of languages sharing a number of parameters constraining the clausal architecture and clitic ordering. The constituting property of both V2 and CL2 syntax is the bottleneck condition licensing only one constituent in XP. The ordering of finite verbs in V2 languages and the ordering of clitics/clitic clusters in CL2 languages is explained as head movement of the diagnostic category (verb or clitic) to 2P accompanied by the XP-movement of phrasal categories. An extension of the class of rigid V2 languages is the class of V1/V2 languages which license IS-motivated V1 orders in declarative clauses, while CL1/CL2 languages are rare.

See All Chapters
Medium 9781601324078

Genetic Algorithms + Evolutionary Strategies and Computations + Optimization Methods

Editied By Hamid R. Arabnia, David de la Fuente, Roger Dziegiel, Elena B. Kozerenko, Peter M. LaMonica,Raymond A. Liuzzi, Jose A. Olivas, Todd Waskiewicz, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti (Editors) CSREA Press PDF

Int'l Conf. Artificial Intelligence | ICAI'15 |

69

Bound Smoothing with a Biased Random-Key Genetic Algorithm

Thiago Alves de Queiroz1 , Ivan da Silva Sendin2 , and Marcos Aur´elio Batista3 of Mathematics and Technology, Federal University of Goi´as - Campus Catal˜ao, Catal˜ao, Goi´as, Brazil.

E-mail: taq@ufg.br.

2 Faculty of Computing, Federal University of Uberlˆ andia, Uberlˆandia, Minas Gerais, Brazil.

E-mail: sendin@ufu.br.

3 Department of Computer Science, Federal University of Goi´ as - Campus Catal˜ao, Catal˜ao, Goi´as, Brazil.

E-mail: marcos.batista@catalao.ufg.br.

1 Institute

Abstract— In this work we solve a subproblem of the distance geometry problem in molecular conformation. The latter aims to determine the three-dimensional structure of a molecule from a set of imprecisely distances. We are interested only in the bound smoothing subproblem, which aims to tighten bounds from a set of lower and upper bounds on distance for pair of atoms. We apply a Biased

Random-Key Genetic Algorithm to solve this subproblem, where each entry of chromosomes indicates how to decrease a bound, with the fitness function measuring the violation of the triangle inequalities. Experimental results show the effectiveness of this algorithm to solve randomly generated instances for which real distances between atoms are known in advance.

See All Chapters
Medium 9781601324078

Artificial Intellegence: Theory, Algorithms and Applications + Cognitive Science + Modeling

Editied By Hamid R. Arabnia, David de la Fuente, Roger Dziegiel, Elena B. Kozerenko, Peter M. LaMonica,Raymond A. Liuzzi, Jose A. Olivas, Todd Waskiewicz, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti (Editors) CSREA Press PDF

Int'l Conf. Artificial Intelligence | ICAI'15 |

407

Visual Intelligence: Toward Machine

Understanding of Video Content

Michael C. Burl, Russell L. Knight, Anthony C. Barrett

Jet Propulsion Laboratory, California Institute of Technology

4800 Oak Grove Drive, Pasadena, CA 91109

Abstract - This paper describes progress toward developing visual intelligence algorithms (VI) that can produce humanlike text descriptions (captions) from video inputs. Video frames are assumed to be generated according to an underlying “script’ that specifies a camera model and the content and action in a scene. VI is formulated as the problem of recovering the script (or relevant portions of the script) given a sequence of video frames. Three types of scripts at different levels of abstraction are recovered: C-scripts contain object detections, poses, and descriptive information on a frame-by-frame basis; B-scripts assign persistent IDs to objects across frames and “smooth” frame-by-frame information; A-scripts provide a symbolic representation of video content using a sparse timeline in which Planning

See All Chapters
Medium 9781601324078

Machine Translation, Natural Language Processing and Related Methods

Editied By Hamid R. Arabnia, David de la Fuente, Roger Dziegiel, Elena B. Kozerenko, Peter M. LaMonica,Raymond A. Liuzzi, Jose A. Olivas, Todd Waskiewicz, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti (Editors) CSREA Press PDF


WorldComp is an
international conference that serves researchers, scholars, professionals,
students, and academicians who are looking to both foster working
relationships and gain access to the latest research results. It is being
held jointly (same location and dates) with a number of other research
conferences; namely, The 2015 World Congress in Computer Science, Computer
Engineering, and Applied Computing (WORLDCOMP'15). The Congress is among the
top five largest annual gathering of researchers in computer science,
computer engineering and applied computing. We anticipate to have attendees
from about 85 countries/territories.

See All Chapters

See All Slices