Notas sobre Sentiment Analysis y Opinion Mining

Existe una asociación enfocada en la promoción de trabajos de procesamiento de lenguaje natural y la lingüística computacional. Se llama: Association for Computational Linguistics, tienen una serie de conferencias, asociaciones regionales y un journal (Computational Lingüistics).

Las conferencias más recientes de la ACL son:

• Annual Meeting of the Association for Computational Linguistics
• Long Papers:
• Cross-domain co-extraction of sentiment and topic lexicons
• Cross-lingual mixture model for sentiment classification
• Polarity consistency checking for sentiment dictionaries
• Short Papers
• Baselines and Bigrams: Simple, Good Sentiment and Topic Classification
• Identifying High-Impact Sub-Structures for Convolution Kernels in Document-level Sentiment Classification
• International Conference on Intelligent Text Processing and Computational Linguistics (CICling)
• Feature Specific Sentiment Analysis for Mixed Product Reviews
• Biographies or Blenders: Which Resource is Best for Cross-Domain Sentiment Analysis?
• A Generate-and-Test Method of Detecting Negative-Sentiment Sentences
• Role of Event Actors and Sentiment Holders in Identifying Event-Sentiment Association
• Applying Sentiment and Social Network Analysis in User Modeling
• The 5W Structure for Sentiment Summarization-Visualization-Tracking
• Naive Bayes Classifiers in opinion mining applications. In search of the best feature set
• A Domain Independent Framework to Extract and Aggregate Analogous Features in Reviews
• Learning Lexical Subjectivity Strength for Chinese Opinionated Sentence Identification
• Building Subjectivity Lexicon(s) From Scratch For Essay Data
• Emotion Ontology Construction From Chinese Knowledge
• Complementary Proceedings:
• Harnessing Wordnet Senses to Unify Sentiment Across Languages
• Summarizing Public Opinions in Tweets
• International Conference on Computational Linguistics (COLing)
• Automatic Detection of Point of View Differences in Wikipedia
• Sentiment Analysis in Twitter with Lightweight Discourse Analysis
• A System For Multilingual Sentiment Learning On Big Data
• Multi-View AdaBoost for Multilingual Subjectivity Analysis
• Finding Thoughtful Comments from Social Media
• Extraction of Russian Sentiment Lexicon for Product Meta-Domain
• International Joint Conference on Natural Language Processing
• Full papers
• A POS-based Ensemble Model for Cross-domain Sentiment Classification
• Automatic Topic Model Adaptation for Sentiment Analysis in Structured Domains
• Detecting and Blocking False Sentiment Propagation
• Enhancing the HL-SOT Approach to Sentiment Analysis via a Localized Feature Selection Framework
• Towards Fine-Grained Sentiment Analysis with Structural Features
• Short papers
• Mining the Sentiment Expectation of Nouns Using Bootstrapping Method
• Posters
• Discovering Resource Terms for Sentiment Analysis
• Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning
• Active Learning for Imbalanced Sentiment Classification
• Why Question Answering using Sentiment Analysis and Word Classes
• Do Neighbours Help? An Exploration of Graph-based Algorithms for Cross-domain Sentiment Classification
• Opinion Target Extraction Using Word-Based Translation Model
• Answering Product Opinionated Questions by Exploiting Hierarchical Organization of Consumer Reviews
• Extracting Opinion Expressions with semi-Markov Conditional Random Fields
• Recent Advances in Natural Language
• Sentiments and Opinions in Health-related Web messages
• An Exploration into the Use of Contextual Document Clustering for Cluster Sentiment Analysis
• Evaluating the Robustness of EmotiBlog for Sentiment Analysis and Opinion Mining
• Multilingual Entity-Centered Sentiment Analysis Evaluated by Parallel Corpora

En general, la ACL es una agrupación que consolida diversas fuentes, las conferencias presentadas cuentan con el apoyo (patrocinio) de empresas como Google, IBM, Microsoft, etc. Todos los artículos presentados se encuentran en su repositorio en línea: http://aclweb.org/anthology-new/

Otras conferencias con artículos publicados relacionados a Sentiment Analysis y Opinion Mining son:

• ACM International Conference on Web Search and Data Mining
• A Large-Scale Sentiment Analysis for Yahoo! Answers
• Mining Contrastive Opinions on Political Texts using the Cross-perspective Topic Model
• Find Me Opinion Sources in Blogosphere: Opinionated Blog Feed Retrieval
• Mining Contrastive Opinions on Political Texts using the Cross-perspective Topic Model
• International Word Wide Web Conference:
• Micropinion Generation: An Unsupervised Approach to Generating Ultra-Concise Summaries of Opinions
• Building reputation and trust using federated search and opinion mining
• Automatic Construction of a Context-Aware Sentiment Lexicon: An Optimization Approach
• ACM Conference on Knowledge Discovery and Data Mining (SIGKDD)
• Entity-Centric Topic-Oriented Opinion Summarization in Twitter
• From Bias to Opinion: A Transfer-Learning Approach to Sentiment Analysis
• International conference on Applications of Natural Language Processing to Information Systems
• Blog Distillation via Sentiment-Sensitive Link Analysis
• Improving document-level sentiment classification using contextual valence shifters
• Extracting Social Events Based on Timeline and Sentiment analysis in Twitter Corpus
• Comparing Different Methods for Opinion Mining in Newspaper Articles
• Can Text Summaries help to Predict Ratings? A Case Study of Movie Reviews

Algunas de las áreas de investigación son:

• Expandir el uso a que no sea sólo un tema.
• Expandir/complementar el uso con textos de distintos idiomas.
• Creación de un perfil (modelo) de usuario en base a sus opiniones.
• Pre-procesamiento (creación de diccionarios, descubrimiento de términos, etc.)
• Aplicaciones en un idioma en específico (sub-problemas basados en el idioma).
• Predicción de aceptación de un producto/tema.
• Optimización de parámetros.