10th National Natural Language Processing Research Symposium

10NNLPRS Pre-Conference Activity

CSP SIG-NLP Membership Launching for Viz-Min and NLP Research Talks
January 24, 2014, Friday
University of Cebu
Cebu, Philippines

Natural Language Processing (NLP) is a field of research that deals with analyzing, understanding and generating the languages that humans use. It combines insights from linguistics and computer science to produce applications like machine translation, information retrieval, and spell checking; all these applications pose strong future impact to business, education, and the wider society in general.

These series of talks aim to train both faculty and students in the current trends and advances in NLP, and in exploring the challenges and prospects of conducting NLP research in academic settings. The talks are co-hosted by the Computing Society of the Philippines-Special Interest Group in Natural Language Processing and the Center for Language Technologies of De La Salle University. The local host is the University of Cebu.

Specific objectives are:

1. To establish as a mechanism for leveraging NLP research in academic institutions;
2. To provide the faculty (and students) of participating academic institutions with orientation on NLP-based research and trainings on the use of various NLP tools; and
3. To provide an avenue for faculty (and students) of academic institutions to get assistance in the conduct of NLP-based researches.

NLP Research Talks

08:30 – 08:55	Registration
08:55 – 09:00	VizMin CSP SIG-NLP Membership Launching	Dr. Rachel Edita O. Roxas
09:00 – 10:00	Toward an Optimal Multilingual Natural Language Generator: Deeper Source Analysis and Shallower Target Analysis	Dr. Tod J. Allman
10:00 – 11:00	NLP Research Towards a Progressive Philippines	Dr. Rachel Edita O. Roxas
11:00 – 12:00	Filipino Sentiment Analysis	Ms. Charibeth K. Cheng
12:00 – 13:00	Lunch
13:00 – 14:00	Attempting to Make Sense out of Social Media Activity	Mr. Leif Romeritch L. Syliongka DLSU
14:00 – 16:00	Who is Walter White? An Introduction to Classification	Mr. Nathaniel A. Oco DLSU

ABOUT THE SPEAKERS

---

Tod Allman (Ph.D. & MA in Linguistics, and an MS and BS in Engineering) is an Adjunct Professor, Department of Applied Linguistics, Graduate Institute of Applied Linguistics, Dallas, TX. He has been working in the field of Natural Language Generation for the past twenty years. He and his colleagues have designed and developed a linguistically based natural language generator named Translator’s Assistant (TA). TA produces high quality draft translations in a wide variety of languages, particularly minority and endangered languages. Linguists may use TA to simultaneously document a language, and also produce initial draft translations of significant texts in the language. When experienced mother-tongue translators edit the drafts produced by this system into publishable texts, their productivity is typically quadrupled without any loss of quality. TA incorporates extensive typological, semantic, syntactic, and discourse research into its semantic representational system and its transfer and synthesizing grammars. Tod has worked with linguists and mother-tongue speakers in order to develop lexicons and grammars for a variety of languages including Korean, Kewa (Papua New Guinea), Jula (Cote d’Ivoire), Angas (Nigeria), and Chinantec (Mexico). He hopes that the texts generated by TA will empower the speakers of these languages by providing them with vital information which helps them live longer, healthier, and more productive lives, and also enables them to participate in the larger world.

Toward an Optimal Multilingual Natural Language Generator: Deeper Source Analysis and Shallower Target Analysis

 

Translator’s Assistant (TA) is a large scale multilingual natural language generator (NLG) designed and developed entirely from a linguist’s perspective. TA has been tested with English, Korean, Kewa (Papua New Guinea), and Jula (Cote d’Ivoire), and proof of concept grammars have been developed for a variety of other languages. TA has generated initial draft translations of texts in each of the test languages, and when experienced mother-tongue translators edit those drafts into publishable texts, their productivity is typically quadrupled when compared with manual translation.

An optimal NLG will be able to generate high quality texts in a wide variety of languages with minimal knowledge of the target language grammars. In order to increase the quality of the drafts generated by TA, deep source analysis techniques have been adopted. And in order to minimize the target language knowledge that is required to generate the drafts, a new approach to grammar development has been designed into LA’s synthesizing grammar. This paper will describe several of the source analysis conventions that have been adopted during the development of TA’s semantic representations, and it will also present examples of the new type of synthesizing rule that was added to TA’s grammar. The adoption of deep source analysis techniques combined with shallow target analysis has proven to be a very efficient model.

---

Dr. Rachel Edita Roxas is a senior researcher of the College of Computer Studies, De La Salle University, Manila, the Philippines. Dr. Roxas obtained her Ph.D. in Computer Science from the Australian National University and her B.S. in Computer Science cum laude from the University of the Philippines at Los Baños. She is the elected founding president of the special interest group on NLP of the Computing Society of the Philippines (CSP SIG-NLP) and has served in various program committees of both local and international conferences on NLP. She is a member of the Steering Committee of the Pacific Asia Conference on Language, Information, and Computation (PACLIC) as a country representative of the Philippines. Her interests include the use of Natural Language Processing (or NLP) in applications in governance, business, education, and others. She is the project leader of the eParticipation/eLegislation component of the PAN eGov program.

NLP Research towards a Progressive Philippines

 

Dr. Roxas will describe some of her projects in NLP that contribute to the advancement of the nation. These include the multi-lingual corpus on Philippine languages, eParticipation and ASEAN-MT project. She will also describe how the other Philippine institutions both students and faculty can contribute to these endeavours.

 

---

Ms. Charibeth K. Cheng is an Assistant Professor of the Software Technology Department of the College of the Computer Studies at De La Salle University. She is interested in developing efficient algorithms to process text written in Filipino to make the language understandable to the computer. She is currently working on a system that determines the subjective information found in a Filipino text.

Filipino Sentiment Analysis

What is the general sentiment of the people about the president? What is the stand of each senator about a particular issue? Do people prefer Brand X over Brand Y? Which features do people look for in a particular product? What do people say about the features of a specific product?

These are just some of the questions considered in sentiment analysis. Sentiment analysis is an area in natural language processing which aims to “understand” the opinion expressed in a text. Among the research on sentiment analysis are, but not limited to, determining whether the sentiment in a text is positive or negative; ranking the sentiment within a range (1 start to 5 stars); identifying the target of the opinion (e.g. poor battery life); comparing opinions; and summarizing reviews.

This talk will present the challenges involved in performing sentiment analysis on Filipino text. What aspects in Filipino make the task difficult? Can we adopt and adapt existing solutions from other languages?

---

Mr. Leif Romeritch Syliongka is a full time faculty of the College of Computer Studies, De La Salle University currently specializing in Ontology, Big Data and Data Mining. He obtained his Bachelor’s degree in Computer Science from De La Salle University – Manila and is currently taking up his Master’s degree in Computer Science in the same university. Recently, he and his team mates entered the Big Data Programming Competition by Trend Micro and was one of the top 10 country finalists.

 

Attempting to Make Sense out of Social Media Activity

With the Philippines being recently labelled as the “Social Media Capital of the World” with over 9.5 Million Twitter Users [1] and 30 Million Facebook Users [2], can you imagine how much posts, likes, follows, friends, or in a nutshell, data, is being poured into these social networking sites everyday by users? Just to give you an idea, worldwide, 58 Millions tweets are generated per day as of May 2013 [3]. The question now is how do you make sense out of these data? How do you turn these data into information? Mr. Syliongka will discuss some of the solutions he and his co-researchers have developed to answer these questions.

1. http://technology.inquirer.net/15189/philippines-has-9-5m-twitter-users-ranks-10th
2. http://www.pulyetos.com/what-we-do/social-media-in-the-philippines-by-the-numbers/
3. http://www.statisticbrain.com/twitter-statistics/

---

Mr. Nathaniel Oco is a researcher of the Center for Language Technologies. He is a member of the Computing Society of the Philippines – Special Interest Group on Natural Language Processing and has served as subreviewer in conferences such as the Philippine Computing Science Congress (PCSC), and the Pacific Asia Conference on Language, Information, and Computation (PACLIC). His research interests lie in natural language processing particularly in rule-based systems and language modeling. He hopes to apply classification techniques and digital signal processing to answer this question someday: “What does the fox say?”

Who is Walter White? An Introduction to Classification

Classification is the process of identifying using computational means which category an input is in. In this lecture, Mr. Oco will discuss classification in general, its mathematical simplicities, useful tools and techniques, and evaluation methodologies. As possible application, he will show how classification can be used to determine which movie a character name is from, and which language a sentence is in.

---

 

10th National Natural Language Processing Research Symposium

https://www.dlsu.edu.ph/conferences/nlp/2014/
February 21-22, 2014, De La Salle University
Co-organized by the Center for Language Technologies, De La Salle University and the Philippine (DLSU), and the Computing Society of the Philippines-Special Interest Group on Natural Language Processing (CSP SIG-NLP).

Organizers

Center for Language Technologies, De La Salle University

The Center for Language Technologies (CeLT) of the College of Computer Studies is dedicated in the pursuit of researches that deal with both programming and human languages.

Computing Society of the Philippines – Special Interest Group on Natural Language Processing

The Computing Society of the Philippines, Special Interest Group – Natural Language Processing (CSP, SIG-NLP) is a group of dedicated NLP researchers from various academic institutions who have joined together in fostering NLP research in the Philippines.