Huge volumes of opinion-rich data is user-generated in social media at an unprecedented rate, easing the analysis of individual and public sentiments. Sentiment analysis has shown to be useful in probing and understanding emotions, expressions and attitudes in the text. However, the distinct characteristics of social media data present challenges to traditional sentiment analysis. First, social media data is often noisy, incomplete and fast-evolved which necessitates the design of a sophisticated learning model. Second, sentiment labels are hard to collect which further exacerbates the problem by not being able to discriminate sentiment polarities. Meanwhile, opportunities are also unequivocally presented. Social media contains rich sources of sentiment signals in textual terms and user interactions, which could be helpful in sentiment analysis. While there are some attempts to leverage implicit sentiment signals in positive user interactions, little attention is paid on signed social networks with both positive and negative links. The availability of signed social networks motivates us to investigate if negative links also contain useful sentiment signals. In this paper, we study a novel problem of unsupervised sentiment analysis with signed social networks. In particular, we incorporate explicit sentiment signals in textual terms and implicit sentiment signals from signed social networks into a coherent model SignedSenti for unsupervised sentiment analysis. Empirical experiments on two real-world datasets corroborate its effectiveness.

In this paper, we focus on the task of extracting named entities together with their associated sentiment information in a joint manner. Our key observation in such an entity-level sentiment analysis (a.k.a. targeted sentiment analysis) task is that there exists a sentiment scope within which each named entity is embedded, which largely decides the sentiment information associated with the entity. However, such sentiment scopes are typically not explicitly annotated in the data, and their lengths can be unbounded. Motivated by this, unlike traditional approaches that cast this problem as a simple sequence labeling task, we propose a novel approach that can explicitly model the latent sentiment scopes. Our experiments on the standard datasets demonstrate that our approach is able to achieve better results compared to existing approaches based on conventional conditional random fields (CRFs) and a more recent work based on neural networks.

Cross-domain sentiment classification aims to tag sentiments for a target domain by labeled data from a source domain. Due to the difference between domains, the accuracy of a trained classifier may be very low. In this paper, we propose a boosting-based learning framework named TR-TrAdaBoost for cross-domain sentiment classification. We firstly explore the topic distribution of documents, and then combine it with the unigram TrAdaBoost. The topic distribution captures the domain information of documents, which is valuable for cross-domain sentiment classification. Experimental results indicate that TR-TrAdaBoost represents documents well and boost the performance and robustness of TrAdaBoost.

Structural correspondence learning (SCL) is an effective method for cross-lingual sentiment classification. This approach uses unlabeled documents along with a word translation oracle to automatically induce task specific, cross-lingual correspondences. It transfers knowledge through identifying important features, i.e., pivot features. For simplicity, however, it assumes that the word translation oracle maps each pivot feature in source language to exactly only one word in target language. This one-to-one mapping between words in different languages is too strict. Also the context is not considered at all. In this paper, we propose a cross-lingual SCL based on distributed representation of words; it can learn meaningful one-to-many mappings for pivot words using large amounts of monolingual data and a small dictionary. We conduct experiments on NLP&CC 2013 cross-lingual sentiment analysis dataset, employing English as source language, and Chinese as target language. Our method does not rely on the parallel corpora and the experimental results show that our approach is more competitive than the state-of-the-art methods in cross-lingual sentiment classification.

The task of aspect and opinion terms co-extraction aims to explicitly extract aspect terms describing features of an entity and opinion terms expressing emotions from user-generated texts. To achieve this task, one effective approach is to exploit relations between aspect terms and opinion terms by parsing syntactic structure for each sentence. However, this approach requires expensive effort for parsing and highly depends on the quality of the parsing results. In this paper, we offer a novel deep learning model, named coupled multi-layer attentions. The proposed model provides an end-to-end solution and does not require any parsers or other linguistic resources for preprocessing. Specifically, the proposed model is a multi-layer attention network, where each layer consists of a couple of attentions with tensor operators. One attention is for extracting aspect terms, while the other is for extracting opinion terms. They are learned interactively to dually propagate information between aspect terms and opinion terms. Through multiple layers, the model can further exploit indirect relations between terms for more precise information extraction. Experimental results on three benchmark datasets in SemEval Challenge 2014 and 2015 show that our model achieves state-of-the-art performances compared with several baselines.

Visual sentiment analysis, which studies the emotional response of humans on visual stimuli such as images and videos, has been an interesting and challenging problem. It tries to understand the high-level content of visual data. The success of current models can be attributed to the development of robust algorithms from computer vision. Most of the existing models try to solve the problem by proposing either robust features or more complex models. In particular, visual features from the whole image or video are the main proposed inputs. Little attention has been paid to local areas, which we believe is pretty relevant to human's emotional response to the whole image. In this work, we study the impact of local image regions on visual sentiment analysis. Our proposed model utilizes the recent studied attention mechanism to jointly discover the relevant local regions and build a sentiment classifier on top of these local regions. The experimental results suggest that 1) our model is capable of automatically discovering sentimental local regions of given images and 2) it outperforms existing state-of-the-art algorithms to visual sentiment analysis.

If you attend a protest in Washington, D.C., nowadays, better plan on leaving your cellphone at home. That is, unless you want police to confiscate it, mine it for incriminating information and then gather even more data from their BFF -- Facebook. At least one person arrested during protests on Inauguration Day got an email from Facebook's Law Enforcement Response Team alerting them that investigators wanted access to their data. Another received a Facebook data subpoena. The email was basically a countdown to when Facebook inevitably handed that data over to D.C. police. That is, unless the respondent figured out how to file an objection within a 10-day window.

According to the World Bank, risky behaviors are increasingly widespread globally and pose a growing threat to individual health and society. Recently, a number of studies have been done to study risky behaviors, such as understanding illicit drug use behaviors using social media data, and predicting drinking behavior and alcohol-related problems among fraternity and sorority members. However, the majority of the related work only focuses on one risky behavior. Research in clinical psychology and public health domains tell us that there may exist some correlations among risk behaviors. In this paper, in order to support and utilize this correlation, we investigate five risky behaviors: drug consumption, drinking, sleep disorder, depression, and eating disorder. We utilize Instagram data to discover the correlation between those five risk behaviors and employ multi-task machine learning techniques to predict the potential risk behaviors in the near future for the Instagram users.

This blog post is the result of my efforts to show to a coworker how to get the insights he needed by using the streaming capabilities and concise API of Apache Spark. In this blog post, you'll learn how to do some simple, yet very interesting analytics that will help you solve real problems by analyzing specific areas of a social network. Using a subset of a Twitter stream was the perfect choice to use in this demonstration, since it had everything we needed: an endless and continuous data source that was ready to be explored. Spark Streaming is very well explained here and in chapter 6 of the ebook "Getting Started with Apache Spark," so we are going to skip some of the details about the Streaming API and move on to setting up our app. Let's see how to prepare our app before doing anything else.

Natural language processing technologies, which are the basis for sentiment analysis of social media platforms and are deployed in some search engine results, can recognize the intended meanings of terms despite different spellings, diction, connotations, and languages, making integration and analytics efforts more comprehensive. These cognitive computing capabilities are responsible for the parsing of unparalleled quantities of big data in integration and analytics efforts in the healthcare space, facilitating advancements in research and treatment options and testing optimization and enhancing master data management. These capabilities can also incorporate real-time geospatial, weather, news, and industry-specific data to influence marketing, sales, and investment opportunities in any number of verticals. Significantly, natural language processing can also provide explanations for analytics results and recommendations, effectively qualifying quantitative facts.