It is difficult to train a personalized task-oriented dialogue system because the data collected from each individual is often insufficient. Personalized dialogue systems trained on a small dataset can overfit and make it difficult to adapt to different user needs. One way to solve this problem is to consider a collection of multiple users' data as a source domain and an individual user's data as a target domain, and to perform a transfer learning from the source to the target domain. By following this idea, we propose "PETAL"(PErsonalized Task-oriented diALogue), a transfer-learning framework based on POMDP to learn a personalized dialogue system. The system first learns common dialogue knowledge from the source domain and then adapts this knowledge to the target user. This framework can avoid the negative transfer problem by considering differences between source and target users. The policy in the personalized POMDP can learn to choose different actions appropriately for different users. Experimental results on a real-world coffee-shopping data and simulation data show that our personalized dialogue system can choose different optimal actions for different users, and thus effectively improve the dialogue quality under the personalized setting.

Organizations have generated unprecedented amounts of employee feedback through weekly or monthly pulse surveys, annual engagement surveys, and internal social networks and collaboration platforms. But many still struggle with how to efficiently comb through that mountain of information to identify actionable insights leaders can use to improve employee engagement and retention. Some companies are now turning to artificial intelligence (AI) tools to conduct sentiment analysis on employee feedback, gauge how employees feel and address their concerns. While text analysis of survey responses isn't new, the emergence of smarter algorithms enables faster and more precise search and categorization of unstructured data, such as open-ended comments, said Alan Lepofsky, vice president and principal analyst with Constellation Research, a technology research firm in Silicon Valley. Lepofsky, author of the recent report Why Artificial Intelligence Will Power the Future of Work, said vendors have made advances in sentiment analysis technology.

Topic models for text corpora comprise a popular family of methods that have inspired many extensions to encode properties such as sparsity, interactions with covariates, and the gradual evolution of topics. In this paper, we combine certain motivating ideas behind variations on topic models with modern techniques for variational inference to produce a flexible framework for topic modeling that allows for rapid exploration of different models. We first discuss how our framework relates to existing models, and then demonstrate that it achieves strong performance, with the introduction of sparsity controlling the trade off between perplexity and topic coherence.

With Firebase and Cloud Functions, you can easily add machine learning to your app to do translation, sentiment analysis, speech recognition, and computer vision. See all the talks from Google I/O '17 here: https://goo.gl/D0D4VE

Sentiment analysis of tweets is often monolingual and the models provided by machine learning classifiers are usually not applicable across distinct languages. Cross-language sentiment classification usually relies on machine translation strategies in which a source language is translated to the desired target language. Machine translation is costly and the provided results are limited by the quality of the translation that is performed. In this paper, we propose an efficient translation-free deep neural architecture for performing multilingual sentiment analysis of tweets. Our proposed approach benefits from a cost-effective character-based embedding and from optimized convolutions to learn from multiple distinct languages. The resulting model is capable of learning latent features from all languages used during training at once and it does not require any translation process to be performed whatsoever. We empirically evaluate the efficiency and effectiveness of the proposed approach in tweet corpora from four different languages and we show that it presents the best trade-off among four distinct state-of-the-art deep neural architectures for sentiment analysis.

This study examines how differences in corpus size influence the accuracy of Latent Semantic Analysis (LSA) spaces and Latent Dirichlet Allocation (LDA) spaces in two tasks: a word association task and a vocabulary definition test. Specific optimizations were considered in building each semantic model. Initial results indicate that larger corpora lead to greater accuracy and that LDA probabilistic models, similar to LSA vector spaces, can provide insights into cognitive processing at semantic levels.

Text representation in Chinese sentiment analysis is usually working at word or character level. In this paper, we prove that radical-level processing could greatly improve sentiment classification performance. In particular, we propose two types of Chinese radical-based hierarchical embeddings. The embeddings incorporate not only semantics at radical and character level, but also sentiment information. In the evaluation of our embeddings, we conduct Chinese sentiment analysis at sentence level on four different datasets. Experimental results validate our assumption that radical-level semantics and sentiments can contribute to sentence-level sentiment classification and demonstrate the superiority of our embeddings over classic textual features and popular word and character embeddings.

She writes about software development, UI and UX, natural language processing, Big Data, AI, and other IT-related topics.

Many common mistakes can be avoided when testing sentiment data for predictive properties. The term "prediction" is not a legal definition. In assessing the predictive qualities of sentiment data there are no rules for what counts as a signal to be tested for predictive properties with regard to financial assets. However, the method you chose ultimately defines what you mean with the term "prediction". To illustrate the point: Using a more prudent definition of the term, the accuracy in the world's most famous prediction study could have been as low as 47% (7 out of 15) instead of 87% (13 out of 15%). An accuracy rate of 47% would not have produced worldwide media attention and more than 1600 academic citations, in my view.

Natural language processing and machine learning can be applied to student feedback to help university administrators and teachers address problematic areas in teaching and learning. The proposed system analyzes student comments from both course surveys and online sources to identify sentiment polarity, the emotions expressed, and satisfaction versus dissatisfaction. A comparison with direct-assessment results demonstrates the system's reliability.