scholarly journals Online Positive and Unlabeled Learning

2020 ◽  
Author(s):  
Chuang Zhang ◽  
Chen Gong ◽  
Tengfei Liu ◽  
Xun Lu ◽  
Weiqiang Wang ◽  
...  
Keyword(s):  
Online Learning ◽  
Learning Algorithm ◽  
Unlabeled Data ◽  
Sequential Data ◽  
Learning Method ◽  
Pu Learning ◽  
Gradient Based ◽  
Learning Scenarios ◽  
Positive And Unlabeled Learning ◽  
Real World Datasets

Positive and Unlabeled learning (PU learning) aims to build a binary classifier where only positive and unlabeled data are available for classifier training. However, existing PU learning methods all work on a batch learning mode, which cannot deal with the online learning scenarios with sequential data. Therefore, this paper proposes a novel positive and unlabeled learning algorithm in an online training mode, which trains a classifier solely on the positive and unlabeled data arriving in a sequential order. Specifically, we adopt an unbiased estimate for the loss induced by the arriving positive or unlabeled examples at each time. Then we show that for any coming new single datum, the model can be updated independently and incrementally by gradient based online learning method. Furthermore, we extend our method to tackle the cases when more than one example is received at each time. Theoretically, we show that the proposed online PU learning method achieves low regret even though it receives sequential positive and unlabeled data. Empirically, we conduct intensive experiments on both benchmark and real-world datasets, and the results clearly demonstrate the effectiveness of the proposed method.

scholarly journals Positive and Unlabeled Learning via Loss Decomposition and Centroid Estimation

2018 ◽  
Author(s):  
Hong Shi ◽  
Shaojun Pan ◽  
Jian Yang ◽  
Chen Gong
Keyword(s):  
Learning Algorithm ◽  
State Of The Art ◽  
False Negative ◽  
Risk Minimization ◽  
Label Noise ◽  
Hinge Loss ◽  
Pu Learning ◽  
Centroid Estimation ◽  
Positive And Unlabeled Learning ◽  
Real World Datasets

Positive and Unlabeled learning (PU learning) aims to train a binary classifier based on only positive and unlabeled examples, where the unlabeled examples could be either positive or negative. The state-of-the-art algorithms usually cast PU learning as a cost-sensitive learning problem and impose distinct weights to different training examples via a manual or automatic way. However, such weight adjustment or estimation can be inaccurate and thus often lead to unsatisfactory performance. Therefore, this paper regards all unlabeled examples as negative, which means that some of the original positive data are mistakenly labeled as negative. By doing so, we convert PU learning into the risk minimization problem in the presence of false negative label noise, and propose a novel PU learning algorithm termed ?Loss Decomposition and Centroid Estimation? (LDCE). By decomposing the hinge loss function into two parts, we show that only the second part is influenced by label noise, of which the adverse effect can be reduced by estimating the centroid of negative examples. We intensively validate our approach on synthetic dataset, UCI benchmark datasets and real-world datasets, and the experimental results firmly demonstrate the effectiveness of our approach when compared with other state-of-the-art PU learning methodologies.

scholarly journals Multi-Positive and Unlabeled Learning

2017 ◽  
Author(s):  
Yixing Xu ◽  
Chang Xu ◽  
Chao Xu ◽  
Dacheng Tao
Keyword(s):  
Unlabeled Data ◽  
Learning Problem ◽  
Step Method ◽  
Practical Applications ◽  
Pu Learning ◽  
Positive And Unlabeled Learning ◽  
One Step ◽  
Real World Datasets ◽  
Multi Class Classification ◽  
The Given

The positive and unlabeled (PU) learning problem focuses on learning a classifier from positive and unlabeled data. Some methods have been developed to solve the PU learning problem. However, they are often limited in practical applications, since only binary classes are involved and cannot easily be adapted to multi-class data. Here we propose a one-step method that directly enables multi-class model to be trained using the given input multi-class data and that predicts the label based on the model decision. Specifically, we construct different convex loss functions for labeled and unlabeled data to learn a discriminant function F. The theoretical analysis on the generalization error bound shows that it is no worse than k√k times of the fully supervised multi-class classification methods when the size of the data in k classes is of the same order. Finally, our experimental results demonstrate the significance and effectiveness of the proposed algorithm in synthetic and real-world datasets.

scholarly journals Fatigue-Aware Bandits for Dependent Click Models

2020 ◽  
Vol 34 (04) ◽  
pp. 3341-3348
Author(s):  
Junyu Cao ◽  
Wei Sun ◽  
Zuo-Jun (Max) Shen ◽  
Markus Ettl
Keyword(s):  
Online Learning ◽  
Recommender Systems ◽  
Learning Algorithm ◽  
Learning Task ◽  
Discount Factor ◽  
Bandit Problem ◽  
User Fatigue ◽  
Learning Scenarios ◽  
Click Model ◽  
Regret Bound

As recommender systems send a massive amount of content to keep users engaged, users may experience fatigue which is contributed by 1) an overexposure to irrelevant content, 2) boredom from seeing too many similar recommendations. To address this problem, we consider an online learning setting where a platform learns a policy to recommend content that takes user fatigue into account. We propose an extension of the Dependent Click Model (DCM) to describe users' behavior. We stipulate that for each piece of content, its attractiveness to a user depends on its intrinsic relevance and a discount factor which measures how many similar contents have been shown. Users view the recommended content sequentially and click on the ones that they find attractive. Users may leave the platform at any time, and the probability of exiting is higher when they do not like the content. Based on user's feedback, the platform learns the relevance of the underlying content as well as the discounting effect due to content fatigue. We refer to this learning task as “fatigue-aware DCM Bandit” problem. We consider two learning scenarios depending on whether the discounting effect is known. For each scenario, we propose a learning algorithm which simultaneously explores and exploits, and characterize its regret bound.

scholarly journals Exploring Clustering-Based Reinforcement Learning for Personalized Book Recommendation in Digital Library

Information ◽  
2021 ◽  
Vol 12 (5) ◽  
pp. 198
Author(s):  
Xinhua Wang ◽  
Yuchen Wang ◽  
Lei Guo ◽  
Liancheng Xu ◽  
Baozhong Gao ◽  
...  
Keyword(s):  
Reinforcement Learning ◽  
Digital Library ◽  
Learning Algorithm ◽  
State Of The Art ◽  
Decision Making Process ◽  
Learning Method ◽  
Large Collection ◽  
Recommendation Algorithms ◽  
Small Set ◽  
Real World Datasets

Digital library as one of the most important ways in helping students acquire professional knowledge and improve their professional level has gained great attention in recent years. However, its large collection (especially the book resources) hinders students from finding the resources that they are interested in. To overcome this challenge, many researchers have already turned to recommendation algorithms. Compared with traditional recommendation tasks, in the digital library, there are two challenges in book recommendation problems. The first is that users may borrow books that they are not interested in (i.e., noisy borrowing behaviours), such as borrowing books for classmates. The second is that the number of books in a digital library is usually very large, which means one student can only borrow a small set of books in history (i.e., data sparsity issue). As the noisy interactions in students’ borrowing sequences may harm the recommendation performance of a book recommender, we focus on refining recommendations via filtering out data noises. Moreover, due to the the lack of direct supervision information, we treat noise filtering in sequences as a decision-making process and innovatively introduce a reinforcement learning method as our recommendation framework. Furthermore, to overcome the sparsity issue of students’ borrowing behaviours, a clustering-based reinforcement learning algorithm is further developed. Experimental results on two real-world datasets demonstrate the superiority of our proposed method compared with several state-of-the-art recommendation methods.

Construct a biased SVM classifier based on Chebyshev distance for PU learning

2020 ◽  
Vol 39 (3) ◽  
pp. 3749-3767
Author(s):  
Ting Ke ◽  
Min Li ◽  
Lidong Zhang ◽  
Hui Lv ◽  
Xuechun Ge
Keyword(s):  
Iterative Algorithm ◽  
Large Scale ◽  
Learning Algorithm ◽  
Classification Performance ◽  
Support Vector ◽  
Svm Classifier ◽  
Empirical Risk ◽  
Pu Learning ◽  
Sample Points ◽  
Real World Datasets

In some real applications, only limited labeled positive examples and many unlabeled examples are available, but there are no negative examples. Such learning is termed as positive and unlabeled (PU) learning. PU learning algorithm has been studied extensively in recent years. However, the classical ones based on the Support Vector Machines (SVMs) are assumed that labeled positive data is independent and identically distributed (i.i.d) and the sample size is large enough. It leads to two obvious shortcomings. On the one hand, the performance is not satisfactory, especially when the number of the labeled positive examples is small. On the other hand, classification results are not optimistic when datasets are Non-i.i.d. For this reason, this paper proposes a novel SVM classifier using Chebyshev distance to measure the empirical risk and designs an efficient iterative algorithm, named L∞ - BSVM in short. L∞ - BSVM includes the following merits: (1) it allows all sample points to participate in learning to prompt classification performance, especially in the case where the size of labeled data is small; (2) it minimizes the distance of the sample points that are (outliers in Non-i.i.d) farthest from the hyper-plane, where outliers are sufficiently taken into consideration (3) our iterative algorithm can solve large scale optimization problem with low time complexity and ensure the convergence of the optimum solution. Finally, extensive experiments on three types of datasets: artificial Non-i.i.d datasets, fault diagnosis of railway turnout with few labeled data (abnormal turnout) and six benchmark real-world datasets verify above opinions again and demonstrate that our classifier is much better than state-of-the-art competitors, such as B-SVM, LUHC, Pulce, B-LSSVM, NB and so on.

scholarly journals The Factors Affecting Acceptance of E-Learning: A Machine Learning Algorithm Approach

2020 ◽  
Vol 10 (10) ◽  
pp. 270
Author(s):  
Dang-Nhac Lu ◽  
Hong-Quang Le ◽  
Tuan-Ha Vu
Keyword(s):  
Online Learning ◽  
Blended Learning ◽  
Clustering Algorithm ◽  
Learning Algorithm ◽  
Sufficient Conditions ◽  
Learning System ◽  
Support Vector ◽  
Learning Method ◽  
K Nearest Neighbor ◽  
E Learning

The Covid-19 epidemic is affecting all areas of life, including the training activities of universities around the world. Therefore, the online learning method is an effective method in the present time and is used by many universities. However, not all training institutions have sufficient conditions, resources, and experience to carry out online learning, especially in under-resourced developing countries. Therefore, the construction of traditional courses (face to face), e-learning, or blended learning in limited conditions that still meet the needs of students is a problem faced by many universities today. To solve this problem, we propose a method of evaluating the influence of these factors on the e-learning system. From there, it is a matter of clarifying the importance and prioritizing construction investment for each factor based on the K-means clustering algorithm, using the data of students who have been participating in the system. At the same time, we propose a model to support students to choose one of the learning methods, such as traditional, e-learning or blended learning, which is suitable for their skills and abilities. The data classification method with the algorithms multilayer perceptron (MP), random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM) and naïve bayes (NB) is applied to find the model fit. The experiment was conducted on 679 data samples collected from 303 students studying at the Academy of Journalism and Communication (AJC), Vietnam. With our proposed method, the results are obtained from experimentation for the different effects of infrastructure, teachers, and courses, also as features of these factors. At the same time, the accuracy of the prediction results which help students to choose an appropriate learning method is up to 81.52%.

scholarly journals Positive and Unlabeled Learning with Label Disambiguation

2019 ◽  
Author(s):  
Chuang Zhang ◽  
Dexin Ren ◽  
Tongliang Liu ◽  
Jian Yang ◽  
Chen Gong
Keyword(s):  
State Of The Art ◽  
Ground Truth ◽  
Training Data ◽  
Learning Approaches ◽  
Generalization Error ◽  
Binary Classifier ◽  
Learning Problem ◽  
Pu Learning ◽  
Positive And Unlabeled Learning ◽  
Real World Datasets

Positive and Unlabeled (PU) learning aims to learn a binary classifier from only positive and unlabeled training data. The state-of-the-art methods usually formulate PU learning as a cost-sensitive learning problem, in which every unlabeled example is simultaneously treated as positive and negative with different class weights. However, the ground-truth label of an unlabeled example should be unique, so the existing models inadvertently introduce the label noise which may lead to the biased classifier and deteriorated performance. To solve this problem, this paper  proposes a novel algorithm dubbed as "Positive and Unlabeled learning with Label Disambiguation'' (PULD). We first regard all the unlabeled examples in PU learning as ambiguously labeled as positive and negative, and then employ the margin-based label disambiguation strategy, which enlarges the margin of classifier response between the most likely label and the less likely one, to find the unique ground-truth label of each unlabeled example. Theoretically, we derive the generalization error bound of the proposed method by analyzing its Rademacher complexity. Experimentally, we conduct intensive experiments on both benchmark and real-world datasets, and the results clearly demonstrate the superiority of the proposed PULD to the existing PU learning approaches.

Sign in / Sign up

Export Citation Format

Share Document