We propose a new online feature selection framework for applications with streaming features where the knowledge of the full feature space is unknown in advance. We define streaming features as features that flow in one by one over time whereas the number of training examples remains fixed. This is in contrast with traditional online learning methods that only deal with sequentially added observations, with little attention being paid to streaming features. The critical challenges for Online Streaming Feature Selection (OSFS) include 1) the continuous growth of feature volumes over time, 2) a large feature space, possibly of unknown or infinite size, and 3) the unavailability of the entire feature set before learning starts. In the paper, we present a novel Online Streaming Feature Selection method to select strongly relevant and nonredundant features on the fly. An efficient Fast-OSFS algorithm is proposed to improve feature selection performance. The proposed algorithms are evaluated extensively on high-dimensional datasets and also with a real-world case study on impact crater detection. Experimental results demonstrate that the algorithms achieve better compactness and higher prediction accuracy than existing streaming feature selection algorithms.

/pdf/online-feature-selection-with-streaming-features-2tyet096q4.pdf

Online Feature Selection with Streaming Features

Due to complex semantics, a sample may be associated with multiple labels in various classification and recognition tasks. Multilabel learning generates training models to map feature vectors to multiple labels. There are several significant challenges in multilabel learning. Samples in multilabel learning are usually described with high-dimensional features and some features may be sequentially extracted. Thus, we do not know the full feature set at the beginning of learning, referred to as streaming features. In this paper, we introduce fuzzy mutual information to evaluate the quality of features in multilabel learning, and design efficient algorithms to conduct multilabel feature selection when the feature space is completely known or partially known in advance. These algorithms are called multilabel feature selection with label correlation (MUCO) and multilabel streaming feature selection (MSFS), respectively. MSFS consists of two key steps: online relevance analysis and online redundancy analysis. In addition, we design a metric to measure the correlation between the label sets, and both MUCO and MSFS take label correlation to consideration. The proposed algorithms are not only able to select features from streaming features, but also able to select features for ordinal multilabel learning. However streaming feature selection is more efficient. The proposed algorithms are tested with a collection of multilabel learning tasks. The experimental results illustrate the effectiveness of the proposed algorithms.

Streaming Feature Selection for Multilabel Learning Based on Fuzzy Mutual Information

When tackling high dimensionality in data mining, online feature selection which deals with features flowing in one by one over time, presents more advantages than traditional feature selection methods. However, in real-world applications, such as fraud detection and medical diagnosis, the data is high-dimensional and highly class imbalanced, namely there are many more instances of some classes than others. In such cases of class imbalance, existing online feature selection algorithms usually ignore the small classes which can be important in these applications. It is hence a challenge to learn from high-dimensional and class imbalanced data in an online manner. Motivated by this, we first formalize the problem of online streaming feature selection for class imbalanced data, and then present an efficient online feature selection framework regarding the dependency between condition features and decision classes. Meanwhile, we propose a new algorithm of Online Feature Selection based on the Dependency in K nearest neighbors, called K-OFSD. In terms of Neighborhood Rough Set theory, K-OFSD uses the information of nearest neighbors to select relevant features which can get higher separability between the majority class and the minority class. Finally, experimental studies on seven high-dimensional and class imbalanced data sets show that our algorithm can achieve better performance than traditional feature selection methods with the same numbers of features and state-of-the-art online streaming feature selection algorithms in an online manner.

Online feature selection for high-dimensional class-imbalanced data

Online streaming feature selection using adapted Neighborhood Rough Set

http://www.cs.umb.edu/~ding/papers/sdarticle.pdf

Detection of Sub-Kilometer Craters in High Resolution Planetary Images Using Shape and Texture Features

Counting craters in remotely sensed images is the only tool that provides relative dating of remote planetary surfaces. Surveying craters requires counting a large amount of small subkilometer craters, which calls for highly efficient automatic crater detection. In this article, we present an integrated framework on autodetection of subkilometer craters with boosting and transfer learning. The framework contains three key components. First, we utilize mathematical morphology to efficiently identify crater candidates, the regions of an image that can potentially contain craters. Only those regions occupying relatively small portions of the original image are the subjects of further processing. Second, we extract and select image texture features, in combination with supervised boosting ensemble learning algorithms, to accurately classify crater candidates into craters and noncraters. Third, we integrate transfer learning into boosting, to enhance detection performance in the regions where surface morphology differs from what is characterized by the training set. Our framework is evaluated on a large test image of 37,500 × 56,250 m2 on Mars, which exhibits a heavily cratered Martian terrain characterized by nonuniform surface morphology. Empirical studies demonstrate that the proposed crater detection framework can achieve an F1 score above 0.85, a significant improvement over the other crater detection algorithms.

/pdf/subkilometer-crater-discovery-with-boosting-and-transfer-1e5658bku4.pdf

Ricardo Ricardo

Papers

Subkilometer crater discovery with boosting and transfer learning