Search

A well-established financial trading system should well perform in resource allocation, risk management, and sustainability. In this paper, we propose a self-management portfolio system with adaptive association mining for practical applications. The system allocates funds into independent units for risk management, and utilizes association mining and adaptive closing mechanism for resource allocation and sustainability, and adopts a self-management module for monitoring positions. The proposed system boosts the annual return and Sharpe ratio to 9.1% and 0.578 (increased to 2.28 and 2.48 times), and reduces the drawdown risk to 34.6% (decreased to almost half). Furthermore, the system rapidly closes the stock positions to avoid drawdown risk in the bear markets, and gradually increases the stock positions when the market turns into bull. Compared with benchmarks, proposed system outperforms all benchmarks in all measurements and on randomly sampled dataset.

Short term load forecasting (STLF) plays an essential role for reliable and economic operation of a power system. In general, the accuracies of the STLF algorithms are dependent on the representation of the input raw data. This paper proposes to use Bag of Words (BoW) model as the data representation to achieve improved accuracy of STLF. As verified in the experiments, the proposed representation of time series load data can drastically improve various STLF algorithms for a microgrid system --- a rather challenging case for load forecasting due to its smaller capacity and higher randomness.

An anomaly, or outlier, is an object exhibiting differences that suggest it belongs to an as-yet undefined class or category. Early detection of anomalies often proves of great importance because they may correspond to events such as fraud, spam, or device malfunctions. By automating the creation of a ranking or list of deviations, we can save time and decrease the cognitive overload of the individuals or groups responsible for responding to such events. Over the years many anomaly and outlier metrics have been developed. In this paper we propose a clustering-based score ensembling method for outlier detection. Using benchmark datasets we evaluate quantitatively the robustness and accuracy of different ensemble strategies. We find that ensembling strategies offer only limited value for increasing overall performance, but provide robustness by negating the influence of severely underperforming models.

Unsupervised domain adaptation deals with scenarios in which labeled data are available in the source domain, but only unlabeled data can be observed in the target domain. Since the classifiers trained by source-domain data would not be expected to generalize well in the target domain, how to transfer the label information from source to target-domain data is a challenging task. A common technique for unsupervised domain adaptation is to match cross-domain data distributions, so that the domain and distribution differences can be suppressed. In this paper, we propose to utilize the label information inferred from the source domain, while the structural information of the unlabeled target-domain data will be jointly exploited for adaptation purposes. Our proposed model not only reduces the distribution mismatch between domains, improved recognition of target-domain data can be achieved simultaneously. In the experiments, we will show that our approach performs favorably against the state-of-the-art unsupervised domain adaptation methods on benchmark data sets. We will also provide convergence, sensitivity, and robustness analysis, which support the use of our model for cross-domain classification.

Botnets are one of most serious threats in cyber security. Many previous studies have been proposed for botnet detection. Among those approaches, one of main tracks focuses on extracting informative features from network traffic flows. Nevertheless, most features of interest are extracted from the information of a single connection, such as flow duration, flow packet size etc. In this paper, we proposed an novel feature, which is able to detect a long-term behavior of botnets. More specifically, we aim to extract a malware beacon from the periodic communication between bots and bot master. Besides the regular communication pattern, we also explore several types of botnet behavior to leverage the effectiveness of the proposed feature. Our experimental results show that our proposed periodic communication signature could be one of effective features for detecting compromised devices.

Debris flows are natural disasters, with soil mass, rocks, and water traveling down a mountainside slope. Debris flows are extremely dangerous; their occurrence incurs huge losses to life and property. The purpose of this research is to develop debris flow detection and emergency evacuation systems. A bag-of-words model is established for analyzing the features of debris flow events, and an anomaly-detection principal component analysis (PCA) model is proposed to detect debris flow. Using real-time debris flow prediction and monitoring, a stochastic optimization model for evacuation planning is formulated. Case studies of debris flow detection in Shenmu village and Fengchiu, central Taiwan, are conducted. Shenmu village and Fengchiu are areas of high potential debris flow, and each has a population of around 800 people. The results show that combining bag-of-words and anomaly-detection PCA methods could predict 6 out of 8 occurrences of actual events, providing a prediction rate of 75 %. In addition, the models make 13 predictions, and 6 of them are correct, providing a prediction accuracy of 46 %. Optimal parameters (including window size, bag length, filter ratio of training data, and anomaly threshold) of the models are also examined to increase the accuracy of debris flow prediction.

We present a novel domain adaptation approach for solving cross-domain pattern recognition problems, i.e., the data or features to be processed and recognized are collected from different domains of interest. Inspired by canonical correlation analysis (CCA), we utilize the derived correlation subspace as a joint representation for associating data across different domains, and we advance reduced kernel techniques for kernel CCA (KCCA) if nonlinear correlation subspace are desirable. Such techniques not only makes KCCA computationally more efficient, potential over-fitting problems can be alleviated as well. Instead of directly performing recognition in the derived CCA subspace (as prior CCA-based domain adaptation methods did), we advocate the exploitation of domain transfer ability in this subspace, in which each dimension has a unique capability in associating cross-domain data. In particular, we propose a novel support vector machine (SVM) with a correlation regularizer, named correlation-transfer SVM, which incorporates the domain adaptation ability into classifier design for cross-domain recognition. We show that our proposed domain adaptation and classification approach can be successfully applied to a variety of cross-domain recognition tasks such as cross-view action recognition, handwritten digit recognition with different features, and image-to-text or text-to-image classification. From our empirical results, we verify that our proposed method outperforms state-of-the-art domain adaptation approaches in terms of recognition performance.

While the task of visual summarization aims to select representative images from an image collection, we solve a unique problem of style-oriented landmark retrieval and summarization from photographic images of a city. Instead of performing summarization or clustering on landmark images from a city, we allow the user to provide a query input which is not from the city of interest, while the goal is to retrieve and summarize the landmark images with similar style-dependent landmark images, followed by a style-consistent image summarization across landmark categories. As a result, our summarized outputs from various landmarks would exhibit similar image style as that of the query. Our experiments will confirm that the use of our proposed method is able to perform favorably against existing or baseline approaches with improved query-dependent style consistency.

Anomaly detection has been an important research topic in data mining and machine learning. Many real-world applications such as intrusion or credit card fraud detection require an effective and efficient framework to identify deviated data instances. However, most anomaly detection methods are typically implemented in batch mode, and thus cannot be easily extended to large-scale problems without sacrificing computation and memory requirements. In this paper, we propose an online oversampling principal component analysis (osPCA) algorithm to address this problem, and we aim at detecting the presence of outliers from a large amount of data via an online updating technique. Unlike prior principal component analysis (PCA)-based approaches, we do not store the entire data matrix or covariance matrix, and thus our approach is especially of interest in online or large-scale problems. By oversampling the target instance and extracting the principal direction of the data, the proposed osPCA allows us to determine the anomaly of the target instance according to the variation of the resulting dominant eigenvector. Since our osPCA need not perform eigen analysis explicitly, the proposed framework is favored for online applications which have computation or memory limitations. Compared with the well-known power method for PCA and other popular anomaly detection algorithms, our experimental results verify the feasibility of our proposed method in terms of both accuracy and efficiency.

Linear discriminant analysis (LDA) is a popular supervised dimension reduction algorithm, which projects the data into an effective low-dimensional linear subspace while the separation between the projected data from different classes is improved. While this subspace is typically determined by solving a generalized eigenvalue decomposition problem, its high computation costs prohibit the use of LDA especially when the scale and the dimensionality of the data are large. Based on the recent success of least squares LDA (LSLDA), we propose a novel rank-one update method with a simplified class indicator matrix. Using the proposed algorithm, we are able to derive the LSLDA model efficiently. Moreover, our LSLDA model can be extended to address the learning task of concept drift, in which the recently received data exhibit with gradual or abrupt changes in distribution. In other words, our LSLDA is able to observe and model the data distribution changes, while the dependency on outdated data will be suppressed. This proposed LSLDA will benefit applications of streaming data classification or mining, and it can recognize data with newly added class labels during the learning process. Experimental results on both synthetic and real datasets (with and without concept drift) confirm the effectiveness of our propose LSLDA.

Motivated by image reconstruction, sparse representation based classification (SRC) has been shown to be an effective method for applications like face recognition. In this paper, we propose a locality-sensitive dictionary learning algorithm for SRC, in which the designed dictionary is able to preserve local data structure, resulting in improved image classification. During the dictionary update and sparse coding stages in the proposed algorithm, we provide closed-form solutions and enforce the data locality constraint throughout the learning process. In contrast to previous dictionary learning approaches utilizing sparse representation techniques, which did not (or only partially) take data locality into consideration, our algorithm is able to produce a more representative dictionary and thus achieves better performance. We conduct experiments on databases designed for face and handwritten digit recognition. For such reconstruction-based classification problems, we will confirm that our proposed method results in better or comparable performance as state-of-the-art SRC methods do, while less training time for dictionary learning can be achieved.

In this paper, we propose a novel algorithm of Generalized Joint Distribution Adaptation (G-JDA) for heterogeneous domain adaptation (HDA), which associates and recognizes cross-domain data observed in different feature spaces (and thus with different dimensionality). With the objective to derive a domain-invariant feature subspace for relating source and target-domain data, our G-JDA learns a pair of feature projection matrices (one for each domain), which allows us to eliminate the difference between projected cross-domain heterogeneous data by matching their marginal and class-conditional distributions. We conduct experiments on cross-domain classification tasks using data across different features, datasets, and modalities. We confirm that our G-JDA would perform favorably against state-of-the-art HDA approaches.

While domain adaptation (DA) aims to associate the learning tasks across data domains, heterogeneous domain adaptation (HDA) particularly deals with learning from cross-domain data which are of different types of features. In other words, for HDA, data from source and target domains are observed in separate feature spaces and thus exhibit distinct distributions. In this paper, we propose a novel learning algorithm of Cross-Domain Landmark Selection (CDLS) for solving the above task. With the goal of deriving a domain-invariant feature subspace for HDA, our CDLS is able to identify representative cross-domain data, including the unlabeled ones in the target domain, for performing adaptation. In addition, the adaptation capabilities of such cross-domain landmarks can be determined accordingly. This is the reason why our CDLS is able to achieve promising HDA performance when comparing to state-of-the-art HDA methods. We conduct classification experiments using data across different features, domains, and modalities. The effectiveness of our proposed method can be successfully verified.

We propose a novel multiple kernel learning (MKL) algorithm with a group lasso regularizer, called group lasso regularized MKL (GL-MKL), for heterogeneous feature fusion and variable selection. For problems of feature fusion, assigning a group of base kernels for each feature type in an MKL framework provides a robust way in fitting data extracted from different feature domains. Adding a mixed norm constraint (i.e., group lasso) as the regularizer, we can enforce the sparsity at the group/feature level and automatically learn a compact feature set for recognition purposes. More precisely, our GL-MKL determines the optimal base kernels, including the associated weights and kernel parameters, and results in improved recognition performance. Besides, our GL-MKL can also be extended to address heterogeneous variable selection problems. For such problems, we aim to select a compact set of variables (i.e., feature attributes) for comparable or improved performance. Our proposed method does not need to exhaustively search for the entire variable space like prior sequential-based variable selection methods did, and we do not require any prior knowledge on the optimal size of the variable subset either. To verify the effectiveness and robustness of our GL-MKL, we conduct experiments on video and image datasets for heterogeneous feature fusion, and perform variable selection on various UCI datasets.

We introduce a technique to improve iterative kernel principal component analysis (KPCA) robust to outliers due to undesirable artifacts such as noises, alignment errors, or occlusion. The proposed iterative robust KPCA (rKPCA) links the iterative updating and robust estimation of principal directions. It inherits good properties from these two ideas for reducing the time complexity, space complexity, and the influence of these outliers on estimating the principal directions. In the asymptotic stability analysis, we also show that our iterative rKPCA converges to the weighted kernel principal kernel components from the batch rKPCA. Experimental results are presented to confirm that our iterative rKPCA achieves the robustness as well as time saving better than batch KPCA.

Sliced inverse regression (SIR) is a renowned dimension reduction method for finding an effective low-dimensional linear subspace. Like many other linear methods, SIR can be extended to nonlinear setting via the ldquokernel trick.rdquo The main purpose of this paper is two-fold. We build kernel SIR in a reproducing kernel Hilbert space rigorously for a more intuitive model explanation and theoretical development. The second focus is on the implementation algorithm of kernel SIR for fast computation and numerical stability. We adopt a low-rank approximation to approximate the huge and dense full kernel covariance matrix and a reduced singular value decomposition technique for extracting kernel SIR directions. We also explore kernel SIR's ability to combine with other linear learning algorithms for classification and regression including multiresponse regression. Numerical experiments show that kernel SIR is an effective kernel tool for nonlinear dimension reduction and it can easily combine with other linear algorithms to form a powerful toolkit for nonlinear data analysis.

This letter discusses the robustness issue of kernel principal component analysis. A class of new robust procedures is proposed based on eigenvalue decomposition of weighted covariance. The proposed procedures will place less weight on deviant patterns and thus be more resistant to data contamination and model deviation. Theoretical influence functions are derived, and numerical examples are presented as well. Both theoretical and numerical results indicate that the proposed robust method outperforms the conventional approach in the sense of being less sensitive to outliers. Our robust method and results also apply to functional principal component analysis.