Abstract:
There is recent increased interest in automated grammatical error correction, which automatically detects and corrects grammatical errors in essays written by learners of English. In particular, four shared tasks within the natural language processing community were organized in the past four years (HOO 2011, HOO 2012, CoNLL 2013, and CoNLL 2014). In this talk, Prof. Ng will give an overview of the two recent CoNLL shared tasks where he was the lead organizer, and present the findings of the shared tasks. Prof. Ng will review recent approaches that achieve state-of-the-art performance in grammatical error correction, and suggest possible future research directions.

Keynote Speaker: Prof. Hwee Tou Ng
Dr. Hwee Tou Ng is Provost's Chair Professor of Computer Science at the National University of Singapore (NUS) and a Senior Faculty Member at the NUS Graduate School for Integrative Sciences and Engineering. He received a PhD in Computer Science from the University of Texas at Austin, USA. His research focuses on natural language processing and information retrieval. He is a Fellow of the Association for Computational Linguistics (ACL).
He has published papers in premier journals and conferences, including Computational Linguistics, Journal of Artificial Intelligence Research (JAIR), ACM Transactions on Information Systems (TOIS), ACL, EMNLP, SIGIR, AAAI, and IJCAI. His papers received the Best Paper Award at EMNLP 2011 and SIGIR 1997. He is an action editor of the Transactions of the Association for Computational Linguistics (TACL), an editorial board member of Natural Language Engineering, and a steering committee member of ACL SIGNLL.
He has also served as the Editor-in-Chief of ACM Transactions on Asian Language Information Processing (TALIP) (May 2007 − May 2013) and an editorial board member of Computational Linguistics (2004 − 2006) and Journal of Artificial Intelligence Research (JAIR) (Sep 2008 − Aug 2011). He was an elected member of the ACL executive committee (2008 − 2010) and a former secretary of ACL SIGNLL. He was program co-chair of EMNLP 2008, ACL 2005, and CoNLL 2004 conferences, and has served as area chair of ACL, EACL, EMNLP, SIGIR, and AAAI conferences and as session chair and program committee member of many past conferences including ACL, EMNLP, SIGIR, AAAI, and IJCAI.

Abstract:
Over the past few years, many NLP researchers have turned toward Distributional Semantics models and Deep Learning to overcome problems that traditional propositional models of semantics simply cannot handle.
Distributional Semantics (DS) assumes that [much of] the semantics of a word can be captured by a distribution of words associated with it and represented as a vector. But DS models also have shortcomings, noticeably in compositionality: it is unclear how (and whether) one can (or should) compose two vectors to arrive at a third, in such a way that the product also reflects [much of] the semantics of the compound.
Deep learning methods such as Recursive Neural Nets are one way to perform composition that seems to work. But certain phenomena are still beyond the reach of deep learning. Dr. Hovy postulate that decomposing the vector into a tensor, by representing separately the distributions of words associated with specific relations off the target word, provides additional representational power. The result, which we call the Structured Distributional Semantic Model (SDSM), is the basis of a series of experiments on automatically discovering the (latent) relations, building and compressing the tensor space, meaning the similarity between two structured distributed representations, and applying the model to perform wowed similarity detection, coreference resolution, anomaly detection, and other tasks.

Keynote Speaker: Prof. Eduard Hovy
Eduard Hovy is a member of the Language Technology Institute in the School of Computer Science at Carnegie Mellon University. He holds adjunct professorships at universities in China and Canada, and is co-Director of Research for the DHS Center for Command, Control, and Interoperability Data Analytics, a distributed cooperation of 17 universities. Dr. Hovy completed a Ph.D. in Computer Science (Artificial Intelligence) at Yale University in 1987, and was awarded an honorary doctorate from the National Distance Education University (UNED) in Madrid in 2013. From 1989 to 2012 he directed the Human Language Technology Group at the Information Sciences Institute of the University of Southern California.
Dr. Hovy’s research addresses several areas in Natural Language Processing, including machine reading of text, question answering, information extraction, automated text summarization, the semi-automated construction of large lexicons and ontologies, and machine translation. His contributions include the co-development of the ROUGE text summarization evaluation method, the BLANC coreference evaluation method, the Omega ontology, the Webclopedia QA Typology, the FEMTI machine translation evaluation classification, and a model of Structured Distributional Semantics.
Dr. Hovy is the author or co-editor of six books and over 300 technical articles and is a popular invited speaker.
In 2001 Dr. Hovy served as President of the Association for Computational Linguistics (ACL) and in 2001–03 as President of the International Association of Machine Translation (IAMT). Dr. Hovy regularly co-teaches courses and serves on Advisory Boards for institutes and funding organizations in Germany, Italy, Netherlands, and the USA.

Abstract:
Sentiment analysis (SA) or opinion mining is the computational study of people’s opinions, sentiments, attitudes, and emotions. Due to almost unlimited applications and numerous research challenges, SA has been a very active research area in natural language processing (NLP) and data mining. SA is regarded as a semantic analysis problem, but is also highly targeted and bounded because a SA system does not need to fully “understand” a sentence or document. It only needs to comprehend some aspects of its meaning, e.g., positive/negative opinions and their targets.
Due to this targeted nature, it allows us to perform deeper language analyses to gain better insights into NLP than in the general setting because the complexity of the general setting of NLP is just too overwhelming. Thus, although general NL understanding is still far from us, we may be able to solve the SA problem satisfactorily. In this talk, I will first introduce SA and its challenges, and then go into detail to discuss a recent study that aims to solve a SA problem but also contributes significantly to machine learning in the areas of lifelong learning and topic modeling.

Keynote Speaker: Prof. LIU Bing
Bing Liu is a professor of Computer Science at the University of Illinois at Chicago (UIC). He received his PhD in Artificial Intelligence from the University of Edinburgh. Before joining UIC, he was a faculty member at the National University of Singapore. His current research interests include sentiment analysis and opinion mining, data mining, machine learning, and natural language processing (NLP). He has published extensively in top conferences and journals. He is also the author of two books: “Sentiment Analysis and Opinion Mining” (Morgan and Claypool) and “Web Data Mining:
Exploring Hyperlinks, Contents and Usage Data” (Springer). In addition to his research contributions, his work has also made important social impacts.
Some of his work has been widely reported in the press, including a front-page article in The New York Times. On professional services, Liu has served as program chairs of many leading data mining related conferences of ACM, IEEE, and SIAM: KDD, ICDM, CIKM, WSDM, SDM, and PAKDD, as associate editors of several leading data mining journals, e.g., TKDE, TWEB, DMKD, and as area/track chairs or senior technical committee members of numerous NLP, data mining, and Web technology conferences. He currently also serves as the Chair of ACM SIGKDD, and is an IEEE Fellow.

Panel Info:
    We are living in the exciting era where big data, cloud computing, mobile internet and social network joint force to drive new innovations of various kinds to change people’s life and work style. Motivated by this new trend, people have renewed their desire for artificial intelligence (AI). AI is the ability that makes a computer system listen, speak, read, write, search, make decision, answer questions and solve questions. Almost all current great innovations including search engine, “brain”projects, speech assistant and spoken translator are supported by big data and internet and powered by machine learning and deep learning to guide its theory and methodology. AI systems rely on the powerful capability of natural language understanding and generation, which is the core of the theme of NLPCC conference. Therefore, NLPCC 2014 organizes this panel session on The Opportunity and Challenge for AI in Big Data Era during the regular program. In this panel session, we will cover the hot topics about noteworthy progress of AI and NLP, deep learning, knowledge graph and their applications to NLP tasks. The panelist will present their viewpoints from various perspectives on the opportunities and challenges we will face and audience will be invited to have face-to-face discussion with panelist. We hope that this panel session will inspire everybody in the conference to generate thoughtful ideas on new innovations and help to drive the state of the art of the field of NLP and Chinese computing.
    
    The panel list includes:
    Xuanjing Huang Fudan University
    Dekang Lin Google
    Bing Liu University of Illinois at Chicago
    Jianyun Nie University of Montreal