Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
Articles by Xiaolong Wang
Total Records ( 2 ) for Xiaolong Wang
  Waqas Anwar , Xuan Wang , LuLi and Xiaolong Wang
  In this study, we present the preliminary achievement of Hidden Markov Model (HMM) to solve the part of speech tagging problem of Urdu language. The presented HMM is derived from the combination of lexical and transition probabilities. An important feature of our tagger is to combine many distinguished smoothing techniques with HMM model to resolve the data sparseness problem. We note that the proposed HMM based Urdu Part of speech tagger with different smoothing method has achieved significant performance. We evaluate our tagger’s results regarding different smoothing methods and different word level accuracy through Analysis of Variance (ANOVA) and show how present results are significant. Also, we compose a confusion matrix about most frequent error occurring tag pairs. The development of our tagger is an important milestone toward Urdu language processing. This will open some novel research directions to mature Urdu language processing.
  Lu Li , Xuan Wang and XiaoLong Wang
  Conditional maximum entropy models provide a unified framework to integrate arbitrary features from different knowledge sources and have been successfully applied to many natural language processing tasks. Feature selection methods are often used to distinguish good features from bad ones to improve model performance. The selection of features in traditional methods is often performed based on different strategies before or along with feature weight estimation, however, weights themselves should be the only factor to measure the importance of features. This study proposes a new selection method based on divide-and-conquer strategies and well-trained feature spaces of small sizes. Features are divided into small subsets, on each of which a sub-model is built and its features are judged according to their weights. The final model is constructed based on merged feature space from all sub-models. Experiments on part of speech tagging show that this method is feasible and efficient.
Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility