As a precursor to research about Sentiment Analysis with Text Classifiers (Naive Bayes, Maximum Entropy, SVM), Sentiment Analysis with bag-of-words was done and Positive / Negative Sentiment was detected with an accuracy of 60%. This is when only unigrams are used. This percentage will be much when bigrams or trigrams are used (in a next blog-post). See the results at: part 1: http://tinyurl.com/gnlfqqm

This paper presents a co-regularization based approach to semi-supervised domain adaptation. Our proposed approach (EA) builds on the notion of augmented space (introduced in EASYADAPT (EA) [1]) and harnesses unlabeled data in target domain to further enable the transfer of information from source to target. This semi-supervised approach to domain adaptation is extremely simple to implement and can be applied as a pre-processing step to any supervised learner. Our theoretical analysis (in terms of Rademacher complexity) of EA and EA show that the hypothesis class of EA has lower complexity (compared to EA) and hence results in tighter generalization bounds. Experimental results on sentiment analysis tasks reinforce our theoretical findings and demonstrate the efficacy of the proposed method when compared to EA as well as a few other baseline approaches.

While domain adaptation has been actively researched, most algorithms focus on the single-source-single-target adaptation setting. In this paper we propose new generalization bounds and algorithms under both classification and regression settings for unsupervised multiple source domain adaptation. Our theoretical analysis naturally leads to an efficient learning strategy using adversarial neural networks: we show how to interpret it as learning feature representations that are invariant to the multiple domain shifts while still being discriminative for the learning task. To this end, we propose multisource domain adversarial networks (MDAN) that approach domain adaptation by optimizing task-adaptive generalization bounds. To demonstrate the effectiveness of MDAN, we conduct extensive experiments showing superior adaptation performance on both classification and regression problems: sentiment analysis, digit classification, and vehicle counting.

While designing machine learning based text analytics applications, often, NLP data scientists manually determine which NLP features to use based upon their knowledge and experience with related problems. This results in increased efforts during feature engineering process and renders automated reuse of features across semantically related applications inherently difficult. In this paper, we argue for standardization in feature specification by outlining structure of a language for specifying NLP features and present an approach for their reuse across applications to increase likelihood of identifying optimal features.

GENEVA – Europe's physics lab CERN on Wednesday said it had stopped using a Facebook team-chat application because of concerns about handing over data to the U.S. tech giant. CERN said it wound up its Facebook Workplace account on Jan. 31 after the U.S. firm gave it the choice of either paying to use the service or sharing data. "Losing control of our data was unacceptable," CERN said in a blog on Jan. 28, confirmed to AFP by spokeswoman Anais Rassat on Wednesday. CERN said it started using Workplace when it was offered the service for free in 2016. It said some 1,000 members of the CERN community had created accounts and there were around 150 active users each week.

In a previous article, we discussed about Natural Language Processing and various tools that we have to quickly get our hands dirty in this field. This post will be about trying spaCy, one of the most wonderful tools that we have for NLP tasks in Python. Today's objective is to get us acquainted with spaCy and NLP. We will write some code to build a small knowledge graph that will contain structured information extracted from unstructured text. The entire code for the project can be found at the end of this article.

With enterprises that produce large quantities of data there is a growing need for better enterprise search solutions. With the availability of Lucene, Solr and Elasticsearch over the last 10 years, dealing with the challenges of finding content these solutions help in more ways than you realize. Whether your company needs a solution for sentiment analysis, text analytics or advanced faceted search technologies, Solr and Elasticsearch provide a great solution to meet multiple requirements. Enterprise Content Understanding how important text mining/analytics and search technologies are for current enterprise-level businesses, you only need to look at the volume of data that is created across the multitude of various content creation platforms. Most businesses employ many different internal and external software solutions for everything from accounting to social media marketing and industry specific examples such as autocad for digital drawings and engineering.

The coronavirus illness (nCoV) is now an international public health emergency, bigger than the SARS outbreak of 2003. Unlike SARS, this time around scientists have better genome sequencing, machine learning, and predictive analysis tools to understand and monitor the outbreak. During the SARS outbreak, it took five months for scientists to sequence the virus's genome. However, the first 2019-nCoV case was reported in December, and scientists had the genome sequenced by January 10, only a month later. Researchers have been using mapping tools to track the spread of disease for several years.

Aspect-Based Sentiment Analysis (ABSA) deals with the extraction of sentiments and their targets. Collecting labeled data for this task in order to help neural networks generalize better can be laborious and time-consuming. As an alternative, similar data to the real-world examples can be produced artificially through an adversarial process which is carried out in the embedding space. Although these examples are not real sentences, they have been shown to act as a regularization method which can make neural networks more robust. In this work, we apply adversarial training, which was put forward by Goodfellow et al. (2014), to the post-trained BERT (BERT-PT) language model proposed by Xu et al. (2019) on the two major tasks of Aspect Extraction and Aspect Sentiment Classification in sentiment analysis. After improving the results of post-trained BERT by an ablation study, we propose a novel architecture called BERT Adversarial Training (BAT) to utilize adversarial training in ABSA. The proposed model outperforms post-trained BERT in both tasks. To the best of our knowledge, this is the first study on the application of adversarial training in ABSA.

Kremlin analysts could have used Twitter as a source of military intelligence to inform their actions in the 2014 Russia–Ukraine conflict, a study has found. University of California experts showed that location-tagged tweets by Ukraine residents could have been used to map out sentiments towards Russia in real-time. The map they made of pro-Kremlin regions turned out to bear a striking resemblance to the actual areas to which Russia dispatched its special forces. Specifically, this included Crimea and regions in the far east of Ukraine -- where the incoming forces would have been most likely to be seen as liberators. In contrast, the data could also reveal those areas where dispatching forces would have lead to greater resistance and corresponding casualties and costs.