As the emergence and the thriving development of social networks, a huge number of short texts are accumulated and need to be processed. Inferring latent topics of collected short texts is useful for understanding its hidden structure and predicting new contents. Unlike conventional topic models such as latent Dirichlet allocation (LDA), a biterm topic model (BTM) was recently proposed for short texts to overcome the sparseness of document-level word co-occurrences by directly modeling the generation process of word pairs. Stochastic inference algorithms based on collapsed Gibbs sampling (CGS) and collapsed variational inference have been proposed for BTM. However, they either require large computational complexity, or rely on very crude estimation. In this work, we develop a stochastic divergence minimization inference algorithm for BTM to estimate latent topics more accurately in a scalable way. Experiments demonstrate the superiority of our proposed algorithm compared with existing inference algorithms.

On average approximately 40 to 50 percent of tasks in a call center are good candidates for automation. These are tasks that a call center agent or manager can trigger – updating your address, for example. The dialog between the AI and the customer is controlled by how the AI application is programmed and closely measured with human oversight. AI does not run without tight controls in place. The analytics include sentiment analysis that tells management which AI-conducted customer interactions were positive or negative.

The process of liquidity provision in financial markets can result in prolonged exposure to illiquid instruments for market makers. In this case, where a proprietary position is not desired, pro-actively targeting the right client who is likely to be interested can be an effective means to offset this position, rather than relying on commensurate interest arising through natural demand. In this paper, we consider the inference of a client profile for the purpose of corporate bond recommendation, based on typical recorded information available to the market maker. Given a historical record of corporate bond transactions and bond meta-data, we use a topic-modelling analogy to develop a probabilistic technique for compiling a curated list of client recommendations for a particular bond that needs to be traded, ranked by probability of interest. We show that a model based on Latent Dirichlet Allocation offers promising performance to deliver relevant recommendations for sales traders.

Teaching machines to accomplish tasks by conversing naturally with humans is challenging. Currently, developing task-oriented dialogue systems requires creating multiple components and typically this involves either a large amount of handcrafting, or acquiring costly labelled datasets to solve a statistical learning problem for each component. In this work we introduce a neural network-based text-in, text-out end-to-end trainable goal-oriented dialogue system along with a new way of collecting dialogue data based on a novel pipe-lined Wizard-of-Oz framework. This approach allows us to develop dialogue systems easily and without making too many assumptions about the task at hand. The results show that the model can converse with human subjects naturally whilst helping them to accomplish tasks in a restaurant search domain.

One of the core components of modern spoken dialogue systems is the belief tracker, which estimates the user's goal at every step of the dialogue. However, most current approaches have difficulty scaling to larger, more complex dialogue domains. This is due to their dependency on either: a) Spoken Language Understanding models that require large amounts of annotated training data; or b) hand-crafted lexicons for capturing some of the linguistic variation in users' language. We propose a novel Neural Belief Tracking (NBT) framework which overcomes these problems by building on recent advances in representation learning. NBT models reason over pre-trained word vectors, learning to compose them into distributed representations of user utterances and dialogue context. Our evaluation on two datasets shows that this approach surpasses past limitations, matching the performance of state-of-the-art models which rely on hand-crafted semantic lexicons and outperforming them when such lexicons are not provided.

Wherever the winds of the market may blow, he always seems to find a way to deliver impressive returns for his investors and his company, Berkshire Hathaway. Every year he authors his famous "shareholder letter" with his musing about the market and investment strategy and -- perhaps as reflects his continued success -- this sentiment analysis of his letters by data scientist Michael Toth shows that the tone has been generally positive over time. Only five of the forty years of letters show an average negative sentiment: those correspond to market downturns in 1987, 1990, 2001/2002 and 2008. Michael used the R language to generate a sentiment score for each letter, and the process was surprisingly simple (you can find the R code here). The letters are published as PDF documents, from which the text can be extracted using the pdf_text function in the pdftools package.

Back to the University Ranking of my designed application. Ranking technology in my application is to parse tweets crawled from Twitter and then rank related tweets according to their relevance to a specific university. I want to filter high-related tweets (topK) to do the Sentiment Analysis, which will avoid trivial tweets that make our results inaccurate. There are may ranking methods actually, such as rank them based on TF-IDF similarity, text summarization, spatial and temporal factors or machine learning ranking method. Even Twitter itself has provided a method based on time or popularity. However, we need a more advanced method which can filter out the most spam and trivial tweets.

Many social media, like Twitter, Facebook and etc, are evolving to become a source of information for people to scrape varied kinds of data, since microblogs on which users post real time messages shows millions of opinions about their attitudes or sentiment towards hot topics and current issues. Recently, I decided to learn how Regional sentiment analysis can help people to make specific decisions or policy strategies for different regions. Notably, Tweets scraped from Twitter can provide tremendous real-time data for our analysis. In my approach, I develop a Twitter Sentiment Classifier, which will classify a scraped tweet into three main polarities: Positive, Negative and Neutral. To make our analysis more straightforward and clear, I will only extract certain data fields related with one tweet.

Planet Money tries to make a program that reads Donald Trump's tweets and then trades stocks. The first step is training the program to interpret the tweets using something called sentiment analysis.

Representing documents is a crucial component in many NLP tasks, for instance predicting aspect ratings in reviews. Previous methods for this task treat documents globally, and do not acknowledge that target categories are often assigned by their authors with generally no indication of the specific sentences that motivate them. To address this issue, we adopt a weakly supervised learning model, which jointly learns to focus on relevant parts of a document according to the context along with a classifier for the target categories. Derived from the weighted multiple-instance regression (MIR) framework, the model learns decomposable document vectors for each individual category and thus overcomes the representational bottleneck in previous methods due to a fixed-length document vector. During prediction, the estimated relevance or saliency weights explicitly capture the contribution of each sentence to the predicted rating, thus offering an explanation of the rating. Our model achieves state-of-the-art performance on multi-aspect sentiment analysis, improving over several baselines. Moreover, the predicted saliency weights are close to human estimates obtained by crowdsourcing, and increase the performance of lexical and topical features for review segmentation and summarization.