Abstract: Named Entity Recognition (NER) aims to classify each word of a document into predefined target named entity classes and is now-a-days considered to be fundamental for many Natural Language Processing (NLP) tasks such as information retrieval, machine translation, information extraction, question answering systems and others. This paper reports about the development of a NER system for Bengali and Hindi using Support Vector Machine (SVM). Though this state of the art machine learning technique has been widely applied to NER in several well-studied languages, the use of this technique to Indian languages (ILs) is very new. The system makes use of the different contextual information of the words along with the variety of features that are helpful in predicting the four different named (NE) classes, such as Person name, Location name, Organization name and Miscellaneous name. We have used the annotated corpora of 122,467 tokens of Bengali and 502,974 tokens of Hindi tagged with the twelve different NE classes 1, defined as part of the IJCNLP-08 NER Shared Task for South and South East Asian Languages (SSEAL) 2. In addition, we have manually annotated 150K wordforms of the Bengali news corpus, developed from the web-archive of a leading Bengali newspaper. We have also developed an unsupervised algorithm in order to generate the lexical context patterns from a part of the unlabeled Bengali news corpus. Lexical patterns have been used as the features of SVM in order to improve the system performance. The NER system has been tested with the gold standard test sets of 35K, and 60K tokens for Bengali, and Hindi, respectively. Evaluation results have demonstrated the recall, precision, and f-score values of 88.61%, 80.12%, and 84.15%, respectively, for Bengali and 80.23%, 74.34%, and 77.17%, respectively, for Hindi. Results show the improvement in the f-score by 5.13% with the use of context patterns. Statistical analysis, ANOVA is also performed to compare the performance of the proposed NER system with that of the existing HMM based system for both the languages.
Abstract: 4G Communication Networks provide heterogeneous
wireless technologies to mobile subscribers through IP based
networks and users can avail high speed access while roaming across
multiple wireless channels; possible by an organized way to manage
the Quality of Service (QoS) functionalities in these networks. This
paper proposes the idea of developing a novel QoS optimization
architecture that will judge the user requirements and knowing peak
times of services utilization can save the bandwidth/cost factors. The
proposed architecture can be customized according to the network
usage priorities so as to considerably improve a network-s QoS
performance.
Abstract: Air conditioning is mainly use as human comfort
cooling medium. It use more in high temperatures are country such as
Malaysia. Proper estimation of cooling load will archive ideal
temperature. Without proper estimation can lead to over estimation or
under estimation. The ideal temperature should be comfort enough.
This study is to develop a program to calculate an ideal cooling load
demand, which is match with heat gain. Through this study, it is easy
to calculate cooling load estimation. Objective of this study are to
develop user-friendly and easy excess cooling load program. This is
to insure the cooling load can be estimate by any of the individual
rather than them using rule-of-thumb. Developed software is carryout
by using Matlab-GUI. These developments are only valid for
common building in Malaysia only. An office building was select as
case study to verify the applicable and accuracy of develop software.
In conclusion, the main objective has successfully where developed
software is user friendly and easily to estimate cooling load demand.
Abstract: In this research relationship between to be smooth the
interest and output of accepted companies stock at negotiable paper
exchange of Tehran is studied. Static community capacity included
363 companies member of negotiable paper exchange of Tehran that
54 companies were, by considering research limitation, selected from
2004 to 2009. Needed data for model test in librarian method was
chosen from RAH AVARDE NOVIN informative banks, TADBIR
and collecting needed data was selected from Tehran negotiable
paper exchange archive. Given results show that in spite of belief
among people based on companies have more smooth interest have
more output, but resulted outcomes of test-done reveals that there is
no relation between smooth interest and stock output.