Recognizing speech act types in Twitter is of much theoretical interest and practical use. Our previous research did not adequately address the deficiency of training data for this multi-class learning task. In this work, we set out by assuming only a small seed training set and experiment with two semi-supervised learning schemes, transductive SVM and graph-based label propagation, which can l...

In this paper, we introduce a package for semi-supervised learning research in the R programming language called RSSL. We cover the purpose of the package, the methods it includes and comment on their use and implementation. We then show, using several code examples, how the package can be used to replicate well-known results from the semi-supervised learning literature.

Various supervised inference methods can be analyzed as convex duals of a generalized maximum entropy framework, where the goal is to find a distribution with maximum entropy subject to the moment matching constraints on the data. We extend this framework to semi-supervised learning using two approaches: 1) by incorporating unlabeled data into the data constraints and 2) by imposing similarity ...

Graph-based semi-supervised classification heavily depends on a well-structured graph. In this paper, we investigate a mixture graph and propose a method called semi-supervised classification based on mixture graph (SSCMG). SSCMG first constructs multiple k nearest neighborhood (kNN) graphs in different random subspaces of the samples. Then, it combines these graphs into a mixture graph and inc...

Traditional classifiers are trained from labeled data only. Labeled samples are often expensive to obtain, while unlabeled data are abundant. Semi-supervised learning can therefore be of great value by using both labeled and unlabeled data for training. We introduce a semi-supervised learning method named decision-directed approximation combined with Support Vector Machines to detect zones cont...

Applying relational tri-training (R-tri-training for short) to web page classification is investigated in this paper. R-tri-training, as a new relational semi-supervised learning algorithm, is well suitable for learning in web page classification. The semi-supervised component of R-tritraining allows it to exploit unlabeled web pages to enhance the learning performance effectively. In addition,...

Recently, we have proposed several effective Web-based term translation extraction methods exploring Web resources to deal with translation of Web query terms. However, many unknown proper names in Web queries are still difficult to be translated by using our previous Web-based term translation extraction methods. Therefore, in this paper we propose a new hybrid translation extraction method, w...

Random walk plays a significant role in computer science. The popular PageRank algorithm uses random walk. Personalized random walks force random walk to “personalized views” of the graph according to users’ preferences. In this paper, we show the close relations between different preferential random walks and label propagation methods used in semi-supervised learning. We further present a maxi...

Empirical evidence shows that in favorable situations semi-supervised learning (SSL) algorithms can capitalize on the abundance of unlabeled training data to improve the performance of a learning task, in the sense that fewer labeled training data are needed to achieve a target error bound. However, in other situations unlabeled data do not seem to help. Recent attempts at theoretically charact...

In this paper, we discuss the importance of the quality against the quantity of automatically extracted examples for word sense disambiguation (WSD). We first show that we can build a competitive WSD system with a memory-based classifier and a feature set reduced to easily and efficiently computable features. We then show that adding automatically annotated examples improves the performance of ...