Effective sentiment stream analysis with self-augmenting training and demand-driven projection
Published 24 July 2011
Ismael Santana Silva, Janaína Gomide, Adriano Veloso, Wagner Meira, Renato Cordeiro Ferreira
Citations41
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The heart of the approach is a training augmentation procedure which takes as input a small training seed, and then it automatically incorporates new relevant messages to the training data, so that at any given time the model properly reflects the sentiments in the event being analyzed.
Abstract
How do we analyze sentiments over a set of opinionated Twitter messages? This issue has been widely studied in recent years, with a prominent approach being based on the application of classification techniques. Basically, messages are classified according to the implicit attitude of the writer with respect to a query term. A major concern, however, is that Twitter (and other media channels) follows the data stream model, and thus the classifier must operate with limited resources, including labeled data for training classification models. This imposes serious challenges for current classification techniques, since they need to be constantly fed with fresh training messages, in order to track sentiment drift and to provide up-to-date sentiment analysis.
Keywords
Computer Science
Machine LearningSupport-Vector Networks
33,035 Citations1995Corinna Cortes, Vladimir Vapnik
High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated and the performance of the support- vector network is compared to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Mining association rules between sets of items in large databases
14,720 Citations1993Rakesh Agrawal, Tomasz Imieliński +1 more
An efficient algorithm is presented that generates all significant association rules between items in the database of customer transactions and incorporates buffer management and novel estimation and pruning techniques.
Thumbs up?
6,987 Citations2002Bo Pang, Lillian Lee +1 more
This work considers the problem of classifying documents not by topic, but by overall sentiment, e.g., determining whether a review is positive or negative, and concludes by examining factors that make the sentiment classification problem more challenging.
now publishers, Inc. eBooksOpinion Mining and Sentiment Analysis
6,745 Citations2008Bo Pang, Lillian Lee
Foundations and Trends® in Information RetrievalOpinion Mining and Sentiment Analysis
6,204 Citations2008Bo Pang, Lillian Lee
This survey covers techniques and approaches that promise to directly enable opinion-oriented information-seeking systems and focuses on methods that seek to address the new challenges raised by sentiment-aware applications, as compared to those that are already present in more traditional fact-based analysis.
The MIT Press eBooksSemi-Supervised Learning
4,308 Citations2006Olivier Chapelle, Schölkopf, B. +1 more
This first comprehensive overview of semi-supervised learning presents state-of-the-art algorithms, a taxonomy of the field, selected applications, benchmark experiments, and perspectives on ongoing and future research.
Data Mining and Knowledge DiscoveryMining Frequent Patterns without Candidate Generation: A Frequent-Pattern Tree Approach
2,615 Citations2003Jiawei Han, Jian Pei +2 more
A novel frequent-pattern tree (FP-tree) structure is proposed, which is an extended prefix-tree structure for storing compressed, crucial information about frequent patterns, and an efficient FP-tree-based mining method, FP-growth, is developed for mining the complete set of frequent patterns by pattern fragment growth.
Mining high-speed data streams
2,180 Citations2000Pedro Domingos, Geoff Hulten
This paper describes and evaluates VFDT, an anytime system that builds decision trees using constant memory and constant time per example, and applies it to mining the continuous stream of Web access data from the whole University of Washington main campus.
Proceedings of the International AAAI Conference on Web and Social MediaFrom Tweets to Polls: Linking Text Sentiment to Public Opinion Time Series
1,955 Citations2010Brendan O’Connor, Ramnath Balasubramanyan +2 more
This work connects measures of public opinion measured from polls with sentiment measured from text, and finds that temporal smoothing is a critically important issue to support a suc- cessful model.
Mining time-changing data streams
1,725 Citations2001Geoff Hulten, Laurie Spencer +1 more
An efficient algorithm for mining decision trees from continuously-changing data streams, based on the ultra-fast VFDT decision tree learner is proposed, called CVFDT, which stays current while making the most of old data by growing an alternative subtree whenever an old one becomes questionable, and replacing the old with the new when the new becomes more accurate.
Machine LearningLearning in the Presence of Concept Drift and Hidden Contexts
1,432 Citations1996Gerhard Widmer, Miroslav Kubát
A family of learning algorithms that flexibly react to concept drift and can take advantage of situations where contexts reappear are described, including a heuristic that constantly monitors the system's behavior.
Mining concept-drifting data streams using ensemble classifiers
1,254 Citations2003Haixun Wang, Wei Fan +2 more
This paper proposes a general framework for mining concept-drifting data streams using weighted ensemble classifiers, and shows that the proposed methods have substantial advantage over single-classifier approaches in prediction accuracy, and the ensemble framework is effective for a variety of classification models.
A streaming ensemble algorithm (SEA) for large-scale classification
1,199 Citations2001W. Nick Street, YongSeog Kim
A fast algorithm for large-scale or streaming data that classifies as well as a single decision tree built on all the data, requires approximately constant memory, and adjusts quickly to concept drift is presented.
New algorithms for fast discovery of association rules
1,112 Citations1997Mohammed J. Zaki, Srinivasan Parthasarathy +2 more
New algorithms for fast association mining, which scan the database only once, are presented, addressing the open question whether all the rules can be efficiently extracted in a single database pass.
Selecting the right interestingness measure for association patterns
861 Citations2002Pang‐Ning Tan, Vipin Kumar +1 more
An overview of various measures proposed in the statistics, machine learning and data mining literature is presented and it is shown that each measure has different properties which make them useful for some application domains, but not for others.
Proceedings of the International AAAI Conference on Web and Social MediaCharacterizing Microblogs with Topic Models
724 Citations2010Daniel Ramage, Susan Dumais +1 more
A scalable implementation of a partially supervised learning model (Labeled LDA) that maps the content of the Twitter feed into dimensions that correspond roughly to substance, style, status, and social characteristics of posts is presented.
Journal of Machine Learning ResearchDynamic Weighted Majority: An Ensemble Method for Drifting Concepts
659 Citations2007J. Zico Kolter, Marcus A. Maloof
It is concluded that *DWM* outperformed other learners that only incrementally learn concept descriptions, that maintain and use previously encountered examples, and that employ an unweighted, fixed-size ensemble of experts.
Lecture notes in computer scienceSentiment Knowledge Discovery in Twitter Streaming Data
580 Citations2010Albert Bifet, Eibe Frank
To deal with streaming unbalanced classes, a sliding window Kappa statistic is proposed for evaluation in time-changing data streams, and a study on Twitter data is performed using learning algorithms for data streams.
New ensemble methods for evolving data streams
556 Citations2009Albert Bifet, Geoffrey Holmes +3 more
A new experimental data stream framework for studying concept drift, and two new variants of Bagging: ADWIN Bagging and Adaptive-Size Hoeffding Tree (ASHT) Bagging are proposed.
Characterizing debate performance via aggregated twitter sentiment
473 Citations2010Nicholas Diakopoulos, David A. Shamma
An analytical methodology and visual representations are developed that could help a journalist or public affairs person better understand the temporal dynamics of sentiment in reaction to the debate video.
Meeting of the Association for Computational LinguisticsUsing Emoticons to reduce Dependency in Machine Learning Techniques for Sentiment Classification
375 Citations2005Jonathon, Read
Issues in evaluation of stream learning algorithms
334 Citations2009João Gama, Raquel Sebastião +1 more
This paper proposes a general framework for assessing predictive stream learning algorithms, and defends the use of Predictive Sequential methods for error estimate - the prequential error.
Accurate decision trees for mining high-speed data streams
296 Citations2003João Gama, Ricardo Rocha +1 more
The VFDT system is extended in two directions: the ability to deal with continuous data and the use of more powerful classification techniques at tree leaves, which can incorporate and classify new information online, with a single scan of the data, in time constant per example.
ACM SIGMOD RecordBOAT—optimistic decision tree construction
288 Citations1999Johannes Gehrke, Venkatesh Ganti +2 more
Micro-blogging as online word of mouth branding
151 Citations2009Bernard J. Jansen, Mimi Zhang +2 more
Research results investigating micro-blogging as a form of online word of mouth branding and the implications for organizations in using micro- blogging as part of their overall marketing strategy and branding campaigns are reported.
BOAT—optimistic decision tree construction
147 Citations1999Johannes Gehrke, Venkatesh Ganti +2 more
A new algorithm (BOAT) for decision tree construction is introduced that improves upon earlier algorithms in both performance and functionality and is the first scalable algorithm with the ability to incrementally update the tree with respect to both insertions and deletions over the dataset.
Active Learning from Data Streams
112 Citations2007Xingquan Zhu, Peng Zhang +2 more
A classifier ensembling based active learning framework which selectively labels instances from data streams to build an accurate classifier to tackle the challenges raised by data streams' dynamic nature.
Categorizing and mining concept drifting data streams
97 Citations2008Peng Zhang, Xingquan Zhu +1 more
This paper categorizes concept drifting into two scenarios: Loose Concept Dr drifting (LCD) and Rigorous Concept Drifting (RCD), and proposes solutions to handle each of them separately, and proposes a new Optimal Weights Adjustment (OWA) method to determine the optimum weight values for classifiers trained from the most recent data chunk.
Tackling concept drift by temporal inductive transfer
88 Citations2006George Forman
A new learning model is developed that transfers induced knowledge through time to benefit future classifier learning tasks and avoids two main problems with existing work in inductive transfer: scalability and the risk of negative transfer.
Classifier and Cluster Ensembles for Mining Concept Drifting Data Streams
85 Citations2010Peng Zhang, Xingquan Zhu +2 more
This paper proposes a new ensemble model which combines both classifiers and clusters together for mining data streams, and presents a label propagation method to infer each cluster's class label by making full use of both class label information from classifiers, and internal structure information from clusters.
Memory issues in frequent itemset mining
58 Citations2004Bart Goethals
A rigorous comparison between current state of the art techniques is given and a new and simple technique is presented, based on sorting the transaction database, resulting in a sometimes more efficient algorithm for frequent itemset mining using less memory.
Elsevier eBooksStreamMiner
55 Citations2004Wei Fan
This work demonstrates StreamMiner, a random decision-tree ensemble based engine to mine data streams, and demonstrates the following two key functionalities: Detecting possible concept-drift on the fly when the trained streaming model is used to classify incoming data streams without knowing the ground truth.
Incremental rule learning with partial instance memory for changing concepts
30 Citations2004Marcus A. Maloof
This paper describes an incremental rule learner with partial instance memory that uses Widmer and Kubat's heuristic to adjust dynamically the window over which it retains and forgets examples, and suggests learners withpartial instance memory converge more quickly to changing target concepts than algorithms that learn solely from new examples.
Lecture notes in computer scienceIncremental Learning with Partial Instance Memory
26 Citations2002Marcus A. Maloof, Ryszard S. Michalski