Boosting is a very successful classification algorithm that produces a linear combination of "weak" classifiers (a.k.a. base learners) to obtain high quality classification models. In this paper we propose a new boosting algorithm where base learners have structure relationships in the functional space. Though such relationships are generic, our work is particularly motivated by the emerging topic of pattern based classification for semi-structured data including graphs. Towards an efficient incorporation of the structure information, we have designed a general model where we use an undirected graph to capture the relationship of subgraph-based base learners. In our method, we combine both L1 norm and Laplacian based L2 norm penalty with Logit loss function of Logit Boost. In this approach, we enforce model sparsity and smoothness in the functional space spanned by the basis functions. We have derived efficient optimization algorithms based on coordinate decent for the new boosting formulation and theoretically prove that it exhibits a natural grouping effect for nearby spatial or overlapping features. Using comprehensive experimental study, we have demonstrated the effectiveness of the proposed learning methods.

Determining anomalies in data streams that are collected and transformed from various types of networks has recently attracted significant research interest.Principal Component Analysis (PCA) is arguably the most widely applied unsupervised anomaly detection technique for networked data streams due to its simplicity and efficiency. However, none of existing PCA based approaches addresses the problem of identifying the sources that contribute most to the observed anomaly, or anomaly localization. In this paper, we first proposed a novel joint sparse PCA method to perform anomaly detection and localization for network data streams. Our key observation is that we can detect anomalies and localize anomalous sources by identifying a low dimensional abnormal subspace that captures the abnormal behavior of data. To better capture the sources of anomalies, we incorporated the structure of the network stream data in our anomaly localization framework. Also, an extended version of PCA, multidimensional KLE, was introduced to stabilize the localization performance. We performed comprehensive experimental studies on four real-world data sets from different application domains and compared our proposed techniques with several state-of-the-arts. Our experimental studies demonstrate the utility of the proposed methods.

With the development of highly efficient graph data collection technology in many application fields, classification of graph data emerges as an important topic in the data mining and machine learning community. Towards building highly accurate classification models for graph data, here we present an efficient graph feature selection method. In our method, we use frequent subgraphs as features for graph classification. Different from existing methods, we consider the spatial distribution of the subgraph features in the graph data and select those ones that have consistent spatial location.We have applied our feature selection methods to several cheminformatics benchmarks. Our method demonstrates a significant improvement of prediction as compared to the state-of-the-art methods.

No abstract

Recent neural network models have achieved impressive performance on sentiment classification in English as well as other languages. Their success heavily depends on the availability of a large amount of labeled data or parallel corpus. In this paper, we investigate an extreme scenario of cross-lingual sentiment classification, in which the low-resource language does not have any labels or parallel corpus. We propose an unsupervised cross-lingual sentiment classification model named multi-view encoder-classifier (MVEC) that leverages an unsupervised machine translation (UMT) system and a language discriminator. Unlike previous language model (LM) based fine-tuning approaches that adjust parameters solely based on the classification error on training data, we employ the encoder-decoder framework of a UMT as a regularization component on the shared network parameters. In particular, the cross-lingual encoder of our model learns a shared representation, which is effective for both reconstructing input sentences of two languages and generating more representative views from the input for classification. Extensive experiments on five language pairs verify that our model significantly outperforms other models for 8/11 sentiment classification tasks.