Absolute Value Inequality SVM for the PU Learning Problem

文本分类是信息检索的关键问题之一.提取更多的可信反例和构造准确高效的分类器是PU(positive andunlabeled)文本分类的两个重要问题.然而,在现有的可信反例提取方法中,很多方法提取的可信反例数量较少,构建的分类器质量有待提高.分别针对这两个重要步骤提供了一种基于聚类的半监督主动分类方法.与传统的反例提取方法不同,利用聚类技术和正例文档应与反例文档共享尽可能少的特征项这一特点,从未标识数据集中尽可能多地移除正例,从而可以获得更多的可信反例.结合SVM 主动学习和改进的Rocchio 构建分类器,并采用改进的TFIDF(term frequency inverse document frequency)进行特征提取,可以显著提高分类的准确度.分别在3 个不同的数据集中测试了分类结果(RCV1,Reuters-21578,20 Newsgoups).实验结果表明,基于聚类寻找可信反例可以在保持较低错误率的情况下获取更多的可信反例,而且主动学习方法的引入也显著提升了分类精度.

Two birds with one stone: Classifying positive and unlabeled examples on uncertain data streams

In positive and unlabeled (PU) learning problems, only positive examples are labeled. Unlabeled data contain both positive and negative examples. Studies show that positive examples of (secondary) diagnoses, and clinical conditions, such as sepsis, are present in unlabeled hospital administrative data, potentially distorting hospital reimbursement systems, and negatively affecting hospitals' revenue and profitability. We investigate whether PU learning is suitable for improving the quality of hospital administrative data. We train three models on 313,434 hospital cases using hospital cost features: two based on the two-step "spy" approach and one using a robust PU learning method. For model evaluation, we rely exclusively on positive examples due to the PU setting. To further assess model performance, we perform an external validity check: We relabel unlabeled sepsis cases, derive new sepsis rates, and compare them to those reported in medical record review studies. All models identify true positives well in unseen data. External validity checks show, however, that only the robust PU learner effectively discriminates between positives and negatives in the unlabeled data, yielding new sepsis rates within the range of sepsis rates reported in medical record review studies. PU learning can improve the quality of hospital administrative data, but its effectiveness depends strongly on the choice of learning approach and classifier. The output of a PU learner can potentially improve hospital reimbursement systems, hospital revenue and profitability management, and sensitivity analyses in healthcare management science, health economics, health services research, and disease surveillance.

Predicting drug-target interaction based on sequence and structure information

Deceptive reviews detection has attracted significant attention from both business and research communities. However, due to the difficulty of human labeling needed for supervised learning, the problem remains to be highly challenging. This paper proposed a novel angle to the problem by modeling PU (positive unlabeled) learning. A semi-supervised model, called mixing population and individual property PU learning (MPIPUL), is proposed. Firstly, some reliable negative examples are identified from the unlabeled dataset. Secondly, some representative positive examples and negative examples are generated based on LDA (Latent Dirichlet Allocation). Thirdly, for the remaining unlabeled examples (we call them spy examples), which can not be explicitly identified as positive and negative, two similarity weights are assigned, by which the probability of a spy example belonging to the positive class and the negative class are displayed. Finally, spy examples and their similarity weights are incorporated into SVM (Support Vector Machine) to build an accurate classifier. Experiments on gold-standard dataset demonstrate the effectiveness of MPIPUL which outperforms the state-of-the-art baselines.

Multi-domain spoken dialog system should be able to detect more than one domain from a user’s utterance. However, it is difficult to train an accurate binary classifier of a domain based on only positive and unlabeled examples. This paper improves hierarchical clustering algorithm to automatically identify reliable negative examples among unlabeled examples. This paper also verifies three linkage criteria that measure the distance between two clusters. In experiments, the proposed method resulted in the highest gain of F1 score compared to the existing methods.

Learning from positive and unlabeled data or PU learning is the setting where a learner only has access to positive examples and unlabeled data. The assumption is that the unlabeled data can contain both positive and negative examples. This setting has attracted increasing interest within the machine learning literature as this type of data naturally arises in applications such as medical diagnosis and knowledge base completion. This article provides a survey of the current state of the art in PU learning. It proposes seven key research questions that commonly arise in this field and provides a broad overview of how the field has tried to address them.

A bagging SVM to learn from positive and unlabeled examples

This paper studies the problem of Positive Unlabeled learning (PU learning), where positive and unlabeled examples are used for training. Naive Bayes (NB) and Tree Augmented Naive Bayes (TAN) have been extended to PU learning algorithms (PNB and PTAN). However, they require user-specified parameter, which is difficult for the user to provide in practice. We estimate this parameter following [2] by taking the selected completely at random assumption and reformulate these two algorithms with this assumption. Furthermore, based on supervised algorithms Averaged One-Dependence Estimators (AODE), Hidden Naive Bayes (HNB) and Full Bayesian network Classifier (FBC), we extend these algorithms to PU learning algorithms (PAODE, PHNB and PFBC respectively). Experimental results on 20 UCI datasets show that the performance of the Bayesian algorithms for PU learning are comparable to corresponding supervised ones in most cases. Additionally, PNB and PFBC are more robust against unlabeled data, and PFBC generally performs the best.

The relevance feedback in information retrieval (IR) is used for query formulation and consists of assessing a sample of retrieved documents. Probably, the most known probabilistic model based on learning from relevance feedback is the so called Robertson and Sparck-Jones' model (RSJ) (S. Robertson and K. Sparck-Jones, 1976; C. van Rijsbergen, 1979), which assigns a value to the indexing terms. The RSJ function can be alternatively seen as a term ranking function which measures the discriminating power of the term, namely the ability of terms to discriminate relevant from non-relevant documents. The process of relevance feedback is similar to that of two-class learning tasks in machine learning (ML). On the basis of a set of positive and negative examples, the learning task aims to select the “optimal hypothesis” which satisfies the data as much as possible and rightly classifies forthcoming data. We address some theoretical issues which arise with the conventional approaches of both IR and ML, suggest a correspondence between IR and ML relatively to the problem of learning from examples. (4 pages)

Supervised learning methods have been recently exploited to learn gene regulatory networks from gene expression data. They consist of building a binary classifier from feature vectors composed by expression levels of a set of known regulatory connections, available in public databases (eg. RegulonDB, TRRD, Transfac, IPA), and using such a classifier to predict new unknown connections. The input to a binary supervised classifier consists normally of positive and negative examples, but usually the only available information are a partial set of gene regulations, i.e. positive examples, and unlabeled data which could include both positive and negative examples. A fundamental challenge is the choice of negative examples from such unlabeled data to make the classifier able to learn from data. We exploit the known topology of a gene network to select such negative examples and show whether such an assumption benefits the performance of a classifier.

The unmanned aerial vehicles (UAVs) have been extensively used in providing intelligence such as target tracking. In our field experiments, a pre-trained deep neural network (DNN) is deployed on the UAV to identify a target from the captured video frames and enable UAV to keep tracking. However, tracking in real time by the DNN requires a lot of computational resources. This motivates us to consider offloading this type of machine learning (ML) tasks to a mobile edge computing (MEC) server. Specifically, we propose a novel hierarchical ML tasks distribution framework for the UAV tracking system, where the UAV is embedded with lower layers of the pre-trained convolutional neural network (CNN) model due to its limited computing capability, while the MEC server with rich computing resources will handle the higher layers of the CNN model. An optimization problem is formulated to minimize the CNN inference delay while taking into account the communications delay, computing time, and ML error. Insights are provided to understand the tradeoff between communications and ML computing in offloading decisions. Numerical results demonstrate the effectiveness of the proposed ML tasks distribution framework with the optimized offloading strategy.

While microscopy diagnosis through supervised learning for image analysis notably contributes to malaria detection, it has limitations. Among its principle challenges is the manual and tiresome process of data annotation for the classification task. The manual annotation of data is prone to inaccuracy defects due to bias, subjectivity and unclear images resulting into many false positives. This is normally due to personal independent judgements that vary from individual microscopists hence summatively affecting the accuracy of the model. In this paper, we seek to investigate the possibility of classifying the negative far examples and the positive near examples from the positives in thick blood smear images for malaria detection. Assessing the classification performance could potentially inform us of the quality of training dataset and guide n selecting the best training dataset for a malaria parasite detection task. We employ the Mean Squared Error (MSE) to distinguish between positive and negative images. We later investigate the performance of the VGG-16 classification model based on how close or far negative examples are from positives. Experimental results showed that negative examples far from the positives produce better results than those near and that the proposed method could potentially be used to reduce false positives and bias in the training data.

Accurately estimating performance accuracy of machine learning classifiers is of fundamental importance in biomedical research with potentially societal consequences upon the deployment of bestperforming tools in everyday life. Although classification has been extensively studied over the past decades, there remain understudied problems when the training data violate the main statistical assumptions relied upon for accurate learning and model characterization. This particularly holds true in the open world setting where observations of a phenomenon generally guarantee its presence but the absence of such evidence cannot be interpreted as the evidence of its absence. Learning from such data is often referred to as positive-unlabeled learning, a form of semi-supervised learning where all labeled data belong to one (say, positive) class. To improve the best practices in the field, we here study the quality of estimated performance in positive-unlabeled learning in the biomedical domain. We provide evidence that such estimates can be wildly inaccurate, depending on the fraction of positive examples in the unlabeled data and the fraction of negative examples mislabeled as positives in the labeled data. We then present correction methods for four such measures and demonstrate that the knowledge or accurate estimates of class priors in the unlabeled data and noise in the labeled data are sufficient for the recovery of true classification performance. We provide theoretical support as well as empirical evidence for the efficacy of the new performance estimation methods.

A new self-paced learning method for privilege-based positive and unlabeled learning