Results 1 - 10
of
13
Improving Twitter Sentiment Analysis with Topic-Based Mixture Modeling and Semi-Supervised Training
"... In this paper, we present multiple ap-proaches to improve sentiment analysis on Twitter data. We first establish a state-of-the-art baseline with a rich fea-ture set. Then we build a topic-based sen-timent mixture model with topic-specific data in a semi-supervised training frame-work. The topic inf ..."
Abstract
-
Cited by 4 (0 self)
- Add to MetaCart
In this paper, we present multiple ap-proaches to improve sentiment analysis on Twitter data. We first establish a state-of-the-art baseline with a rich fea-ture set. Then we build a topic-based sen-timent mixture model with topic-specific data in a semi-supervised training frame-work. The topic information is generated through topic modeling based on an ef-ficient implementation of Latent Dirich-let Allocation (LDA). The proposed sen-timent model outperforms the top system in the task of Sentiment Analysis in Twit-ter in SemEval-2013 in terms of averaged F scores. 1
Exploiting Social Relations and Sentiment for Stock Prediction
"... In this paper we first exploit cash-tags (“$ ” fol-lowed by stocks ’ ticker symbols) in Twitter to build a stock network, where nodes are stocks connected by edges when two stocks co-occur frequently in tweets. We then employ a labeled topic model to jointly model both the tweets and the network str ..."
Abstract
-
Cited by 2 (0 self)
- Add to MetaCart
(Show Context)
In this paper we first exploit cash-tags (“$ ” fol-lowed by stocks ’ ticker symbols) in Twitter to build a stock network, where nodes are stocks connected by edges when two stocks co-occur frequently in tweets. We then employ a labeled topic model to jointly model both the tweets and the network structure to assign each node and each edge a topic respectively. This Semantic Stock Network (SSN) summarizes discussion topics about stocks and stock relations. We fur-ther show that social sentiment about stock (node) topics and stock relationship (edge) topics are predictive of each stock’s market. For predic-tion, we propose to regress the topic-sentiment time-series and the stock’s price time series. Ex-perimental results demonstrate that topic senti-ments from close neighbors are able to help im-prove the prediction of a stock markedly. 1
TWITTER AND FINANCIAL MARKETS
"... Abstract Several researchers (Mitra and Mitra 2011, Tetlock 2007) ..."
(Show Context)
Linked Data-based Social Media Analysis for Stock Market Tracking
"... Abstract—The rising demand of social media and ensuing activities provide a rich data source for opinion mining. The efforts reported in this paper analyze a popular micro-blogging social network for one of the important application domains, i.e., stock market prediction. The novel approach develope ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—The rising demand of social media and ensuing activities provide a rich data source for opinion mining. The efforts reported in this paper analyze a popular micro-blogging social network for one of the important application domains, i.e., stock market prediction. The novel approach developed includes the use of the Semantic Web technologies to more accurately interpret social media streams. The algorithm closely examines the visible public mood vis-a-vis an organization and related entities, namely products and affiliated persons, and considers the effect of these parameters in calculating the sentiment. The entities and their relationships to the organization of interest are retrieved from one of the largest Linked Data repository available online. Furthermore, we assess the algorithm by collecting the public sentiments as well as the intra-day stock prices during our experiments followed by the statistical tests on the collected data. The evaluation takes into account different company profiles according to our heuristics to study any detected correlation in detail. The exercise is repeated with the benchmark method, which disregards the two additional entities we consider. The evaluation considers whether the fluctuating sentiment is reflected in stock prices, and the most suitable time-lag. I.
Predicting Sector Index Movement with Microblogging Public Mood Time Series on Social Issues
"... This paper develops a technique that unfolds public mood on social issues from real-time so-cial media for sector index prediction. We first propose a low-dimensional support vector ma-chine (SVM) classifier using surrounding infor-mation for twitter sentiment classification. Then, we generate publi ..."
Abstract
- Add to MetaCart
(Show Context)
This paper develops a technique that unfolds public mood on social issues from real-time so-cial media for sector index prediction. We first propose a low-dimensional support vector ma-chine (SVM) classifier using surrounding infor-mation for twitter sentiment classification. Then, we generate public mood time series by aggre-gating message-level weighted daily mood (WDM) based on the sentiment classification re-sults. Lastly, we evaluate our method against the real stock index in two kinds of time periods (fluctuating and monotonous) separately using static cross-correlation coefficient (CCF) and dynamic vector auto-regression (VAR). The ex-periments on “food safety ” issue show that the proposed WDM method outperforms the word-level baseline method in predicting stock move-ment, especially during fluctuating period. 1
Reading Documents for Bayesian Online Change Point Detection
, 2015
"... Modeling non-stationary time-series data for making predictions is a challenging but important task. One of the key is-sues is to identify long-term changes accurately in time-varying data. Bayesian On-line Change Point Detection (BO-CPD) algorithms efficiently detect long-term changes without assum ..."
Abstract
- Add to MetaCart
(Show Context)
Modeling non-stationary time-series data for making predictions is a challenging but important task. One of the key is-sues is to identify long-term changes accurately in time-varying data. Bayesian On-line Change Point Detection (BO-CPD) algorithms efficiently detect long-term changes without assuming the Markov property which is vulnerable to local signal noise. We propose a Document based BO-CPD (DBO-CPD) model which automatically detects long-term temporal changes of continuous variables based on a novel dynamic Bayesian analysis which combines a non-parametric regression, the Gaussian Process (GP), with generative models of texts such as news articles and posts on social networks. Since texts often include important clues of signal changes, DBO-CPD enables the accurate prediction of long-term changes accurately. We show that our algorithm outperforms exist-ing BO-CPDs in two real-world datasets: stock prices and movie revenues.
FLORIN – A System to Support (Near) Real-Time Applica- tions on User Generated Content on Daily News
"... In this paper, we propose a system, FLORIN, which provides sup-port for near real-time applications on user generated content on daily news. FLORIN continuously crawls news outlets for articles and user comments accompanying them. It attaches the articles and comments to daily event stories. It iden ..."
Abstract
- Add to MetaCart
(Show Context)
In this paper, we propose a system, FLORIN, which provides sup-port for near real-time applications on user generated content on daily news. FLORIN continuously crawls news outlets for articles and user comments accompanying them. It attaches the articles and comments to daily event stories. It identifies the opinionated con-tent in user comments and performs named entity recognition on news articles. All these pieces of information are organized hierar-chically and exportable to other applications. Multiple applications can be built on this data. We have implemented a sentiment analysis system that runs on top of it. 1.
Big Data Analytics for Development: Events, Knowledge Graphs and Predictive Models
, 2015
"... Volatility in critical socio-economic indices can have a significant negative impact on global development. This thesis presents a suite of novel big data analytics algorithms that operate on unstructured Web data streams to automatically infer events, knowledge graphs and predictive models to under ..."
Abstract
- Add to MetaCart
(Show Context)
Volatility in critical socio-economic indices can have a significant negative impact on global development. This thesis presents a suite of novel big data analytics algorithms that operate on unstructured Web data streams to automatically infer events, knowledge graphs and predictive models to understand, characterize and predict the volatility of socioeconomic indices. This thesis makes four important research contributions. First, given a large volume of diverse un-structured news streams, we present new models for capturing events and learning spatio-temporal char-acteristics of events from news streams. We specifically explore two types of event models in this thesis: one centered around the concept of event triggers and a probabilistic meta-event model that explicitly de-lineates named entities from text streams to learn a generic class of meta-events. The second contribution focuses on learning several different types of knowledge graphs from news streams and events: a) Spatio-temporal article graphs capture intrinsic relationships between different news articles; b) Event graphs characterize relationships between events and given a news query, provide a succinct summary of a time-line of events relating to a query; c) Event-phenomenon graphs that provide a condensed representation of classes of events that relate to a given phenomena at a given location and time; d) Causality testing on
Using Structured Events to Predict Stock Price Movement: An Empirical Investigation
"... It has been shown that news events influ-ence the trends of stock price movements. However, previous work on news-driven stock market prediction rely on shallow features (such as bags-of-words, named entities and noun phrases), which do not capture structured entity-relation informa-tion, and hence ..."
Abstract
- Add to MetaCart
(Show Context)
It has been shown that news events influ-ence the trends of stock price movements. However, previous work on news-driven stock market prediction rely on shallow features (such as bags-of-words, named entities and noun phrases), which do not capture structured entity-relation informa-tion, and hence cannot represent complete and exact events. Recent advances in Open Information Extraction (Open IE) techniques enable the extraction of struc-tured events from web-scale data. We propose to adapt Open IE technology for event-based stock price movement pre-diction, extracting structured events from large-scale public news without manual efforts. Both linear and nonlinear mod-els are employed to empirically investigate the hidden and complex relationships be-tween events and the stock market. Large-scale experiments show that the accuracy of S&P 500 index prediction is 60%, and that of individual stock prediction can be over 70%. Our event-based system out-performs bags-of-words-based baselines, and previously reported systems trained on S&P 500 stock historical data. 1
