Joseph Chang

Student researcher and freelance developer.

• NTCIR-9 Workshop Meeting on Evaluation of Information Access Technologies

  • September 2011 to December 2011
  Team Members: Joseph Chang, Shih-Ting Huang, Chung-chi Huang, Jason Chang

  The NTCIR Workshop is a series of evaluation workshops designed to enhance research in information access technologies including information retrieval, summarization, extraction, question answering, etc.

• HOO Exercise at BEA 2012

  • February 2012 to June 2012
  Team Members: Joseph Chang, Shih-Ting Huang, Jason Chang, MeiHua Chen

  The HOO (Helping Our Own) Exercise is concerned with correcting textual errors, focusses on the correction of preposition and determiner errors in a large collection of non-native speaker texts.

• 0xBench

  • May 2010 to Present
  Team Members: Joseph Chang, Jim Huang, Julian Chu

  0xlab integrates a series of benchmarks for Android system into the comprehensive benchmark suite, 0xbench.
  
  Key Features
  - Fully open source. License terms: Apache Software License (primary), CDDL (partial)
  - Comprehensive benchmarking: from system call (bionic) to Android frameworks
  - Community development process: accept open source contributions for benchmark items and reference results
  
  Coverage
  - C library and system call
  - OpenGL|ES
  - 2D canvas
  - Garbage collection in Dalvik
  - JavaScript engine
  

• Test of English for International Communication (TOEIC)

  • Educational Testing Service (ETS)
  • License scored 950 / 990

1. C++
2. Python
3. Hadoop Cluster Administration
4. Cloud Computing: Designing MapReduce Programs
5. Linux Server Administration
6. Android Application Development
7. Java

• WikiSense: Supersense Tagging of Unknown Named Entities in WordNet Base on Wikipedia

  • Proceedings of the 23rd Pacific Asia Conference on Language, Information, and Computation (PACLIC23)
  • December 3, 2009
  Authors: Joseph Chang, Richard Tzong-Han Tsai, Jason S. Chang

  Abstract. In this paper, we introduce a minimally supervised method for learning to classify named-entity titles in a given encyclopedia into broad semantic categories in an existing ontology. Our main idea involves using overlapping entries in the encyclopedia and ontology and a small set of 30 handed tagged parenthetic explanations to automatically generate the training data. The proposed method involves automatically recognizing whether a title is a named entity, automatically generating two sets of training data, and automatically building a classification model for training a classification model based on textual and non-textual features. We present WikiSense, an implementation of the proposed method for extending the named entity coverage of WordNet by sense tagging Wikipedia titles. Experimental results shows WikiSense achieves accuracy of over 95% and near 80% applicability for all NE titles in Wikipedia. WikiSense cleanly produces over 1.2 million of NEs tagged with broad categories, based on the lexicographers’ files of WordNet, effectively extending WordNet to form a very large scale semantic category, a potentially useful resource for many natural language related tasks.
  Keywords: semantic category, word sense disambiguation, WordNet, Wikipedia

• Minimally Supervised Question Classification and Answering based on WordNet and Wikipedia

  • Proceedings of the 21st Conference on Computational Linguistics and Speech Processing (ROCLING21)
  • September 1, 2009
  Authors: Joseph Chang, Tzu-Hsi Yen, Richard Tzong-Han Tsai

  Abstract. In this paper, we introduce an automatic method for question classification using fine-grained semantic categories in an existing lexical database (i.e., WordNet) as our class tagset. For this, we also constructed a large scale entity supersense category that classifies over 1,581,865 entities to the top 25 WordNet lexical files (supersense) from titles of Wikipedia entry. To show the usefulness of our work, a simple redundancy based open domain QA system that takes the advantage of the large scale semantic category to perform question classification and named entity classification is also constructed. Experimental results show that the proposed methods outperform the baseline of not using question classification.
  
  Keywords: question answering, question classification, semantic category, WordNet, Wikipedia.

• Learning to Find Translations and Transliterations on the Web

  • (to appear in) The 50th Annual Meeting of the Association for Computational Linguistics (ACL)
  • May 17, 2012
  Authors: Joseph Chang, Jason S. Chang, Roger Jyh-Shing Jang

  Abstract. In this paper, we present a new method for learning to finding translations and transliterations on the Web for a given term. The approach involves using a small set of terms and translations to obtain mixed-code snippets from a search engine, and automatically annotating the snippets with tags and features for training a conditional random field model. At runtime,the model is used to extracting translation candidates for a given term. Preliminary experiments and evaluation show our method cleanly combining various features, resulting in a system that outperforms previous work.

• Chinese

  (Native or bilingual proficiency)

• English

  (Native or bilingual proficiency)

Websites:

• Blog

Interests:

Programming, Photography, Music.

Groups and Associations:

• 
  Android Developer Group
• 
  Computational Linguistics
• 
  Distributed Computing & Applications Professional: Hadoop, MapReduce, NoSQL, MongoDB, VLDB, Big Data
• 
  Hadoop Users
• 
  Linking Taiwan

Honors and Awards:

- Third Award out of 160 teams (~10%), National IC/CAD Programming Contest for Graduate and Undergraduate Students, Ministry of Education, 2009. (教育部大專校院 IC/CAD 程式設計競賽)

- Third Place out of 149 teams, National Information Security Contest (The Golden Shield Contest), Industrial Technology Research Institute and Ministry of Education, 2010. (資策會金盾資安競賽)

- Second Place out of 67 teams, The 11th Trend Micro Programming Contest, Trend Micro. (趨勢科技軟體開發競賽)

- First Place, Fun Taipei Mobile Application Development Contest, Taipei City Government. (台北市政府手機軟體開發競賽)