Clustering-Based PU Active Text Classification Method

Positive and unlabeled learning (PU learning) is a significant binary classification task in machine learning; it focuses on training accurate classifiers using positive data and unlabeled data. Most of the works in this area are based on a two-step strategy: the first step is to identify reliable negative examples from unlabeled examples, and the second step is to construct the classifiers based on the positive examples and the identified reliable negative examples using supervised learning methods. However, these methods always underutilize the remaining unlabeled data, which limits the performance of PU learning. Furthermore, many methods require the iterative solution of the formulated quadratic programming problems to obtain the final classifier, resulting in a large computational cost. In this paper, we propose a new method called the absolute value inequality support vector machine, which applies the concept of eccentricity to select reliable negative examples from unlabeled data and then constructs a classifier based on the positive examples, the selected negative examples, and the remaining unlabeled data. In addition, we apply a hyperparameter optimization technique to automatically search and select the optimal parameter values in the proposed algorithm. Numerical experimental results on ten real-world datasets demonstrate that our method is better than the other three benchmark algorithms.

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.

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.

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

The evolution of Semantic Web (SW) depends on the increasing number of ontologies it contains. However, the existing ontologies are diverse in structure and content, because their design standards are different. In order to ensure their knowledge sharing, the correspondences between different ontology entities should be determined, which is called ontology matching. Currently, various ontology matching techniques have been proposed, which makes use of different strategies or methods to improve the quality of alignment. Being enlightened by the success of Semi-supervised Learning (SL) in ontology matching domain, this paper further proposes aN aive Bayesian Classifier (NBC) based SL, and use it to obtain high-quality alignment. In particular, our approach first models the ontology matching issue as a binary classification problem; then the Positive and Unlabeled (PU) training set and the negative examples that are determined by it is constructed, which is used to train the NBC; and finally, with the learned knowledge, the complete ontology alignment is obtained. The testing cases used in the experiment are provided by Ontology Alignment Evolution Initiative (OAEI). Comparing with the existing ontology matching techniques, the results show that the effectiveness of our method.

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.

Automatic concept learning from large scale imbalanced data sets is a key issue in video semantic analysis and retrieval, which means the number of negative examples is far more than that of positive examples for each concept in the training data. The existing methods adopt generally under-sampling for the majority negative examples or over-sampling for the minority positive examples to balance the class distribution on training data. The main drawbacks of these methods are: (1) As a key factor that affects greatly the performance, in most existing methods, the degree of re-sampling needs to be pre-fixed, which is not generally the optimal choice; (2) Many useful negative samples may be discarded in under-sampling. In addition, some works only focus on the improvement of the computational speed, rather than the accuracy. To address the above issues, we propose a new approach and algorithm named AdaOUBoost (Adaptive Over-sampling and Under-sampling Boost). The novelty of AdaOUBoost mainly lies in: adaptively over-sample the minority positive examples and under-sample the majority negative examples to form different sub-classifiers. And combine these sub-classifiers according to their accuracy to create a strong classifier, which aims to use fully the whole training data and improve the performance of the class-imbalance learning classifier. In AdaOUBoost, first, our clustering-based under-sampling method is employed to divide the majority negative examples into some disjoint subsets. Then, for each subset of negative examples, we utilize the borderline-SMOTE (synthetic minority over-sampling technique) algorithm to over-sample the positive examples with different size, train each sub-classifier using each of them, and get the classifier by fusing these sub-classifiers with different weights. Finally, we combine these classifiers in each subset of negative examples to create a strong classifier. We compare the performance between AdaOUBoost and the state-of-the-art methods on TRECVID 2008 benchmark with all 20 concepts, and the results show the AdaOUBoost can achieve the superior performance in large scale imbalanced data sets.

BackgroundLarge language model (LLM) fine-tuning is the process of adjusting out-of-the-box model weights using a dataset of interest. Fine-tuning can be a powerful technique to improve model performance in fields like medicine, where LLMs may have poor out-of-the-box performance. The 2 common fine-tuning techniques are supervised fine-tuning (SFT) and direct preference optimization (DPO); however, little guidance is available for when to apply either method within clinical medicine or health care operations.ObjectiveThis study aims to investigate the benefits of fine-tuning with SFT and DPO across a range of core natural language tasks in medicine to better inform clinical informaticists when either technique should be deployed.MethodsWe use Llama3 8B (Meta) and Mistral 7B v2 (Mistral AI) to compare the performance of SFT alone and DPO across 4 common natural language tasks in medicine. The tasks we evaluate include text classification, clinical reasoning, text summarization, and clinical triage.ResultsOur results found clinical reasoning accuracy increased from 7% to 22% with base Llama3 and Mistral2, respectively, to 28% and 33% with SFT, and then 36% and 40% with DPO (P=.003 and P=.004, respectively). Summarization quality, graded on a 5-point Likert scale, was 4.11 with base Llama3 and 3.93 with base Mistral2. Performance increased to 4.21 and 3.98 with SFT and then 4.34 and 4.08 with DPO (P<.001). F1-scores for provider triage were 0.55 for Llama3 and 0.49 for Mistral2, which increased to 0.58 and 0.52 with SFT and 0.74 and 0.66 with DPO (P<.001). F1-scores for urgency triage were 0.81 for Llama3 and 0.88 for Mistral2, which decreased with SFT to 0.79 and 0.87, and then experienced mixed results with DPO, achieving 0.91 and 0.85, respectively (P<.001 and P>.99, respectively). Finally, F1-scores for text classification were 0.63 for Llama3 and 0.73 for Mistral2, which increased to 0.98 and 0.97 with SFT, and then essentially did not change with DPO to 0.95 and 0.97, respectively (P=.55 and P>.99, respectively). DPO fine-tuning required approximately 2 to 3 times more compute resources than SFT alone.ConclusionsSFT alone is sufficient for simple tasks such as rule-based text classification, while DPO after SFT improves performance on the more complex tasks of triage, clinical reasoning, and summarization. We postulate that SFT alone is sufficient for simple tasks because SFT strengthens simple word-association reasoning, whereas DPO enables deeper comprehension because it is trained with both positive and negative examples, enabling the model to recognize more complex patterns. Ultimately, our results help inform clinical informaticists when to deploy either fine-tuning method and encourage commercial LLM providers to offer DPO fine-tuning for commonly used proprietary LLMs in medicine.

Audio is an important part of multimedia, and it has many useful applications in real life. Audio event classification is a key technology in audio management and application. Supervised audio event classification requires labeling large amounts of samples, while manual labeling is a very time-consuming work. In this paper we propose SVM CKNND , an active learning method for SVM classifier, to deal with the labeling problem in audio event classification. For SVM CKNND , in each iteration, first, a low-confidence region is delimited; then based on KNN, the samples that are more likely to be on the true class boundary are taken as the informative ones; finally, redundancy that exists in the informative samples is reduced to further decrease manual labeling workload. Experimental results show that SVM CKNND performs better than another two SVM active learning algorithms, especially in classifying small-sample audio events.

The problem of PU Learning, i.e., learning classifiers with positive and unlabelled examples (but not negative examples), is very important in information retrieval and data mining. We address this problem through a novel approach: reducing it to the problem of learning classifiers for some meaningful multivariate performance measures. In particular, we show how a powerful machine learning algorithm, support vector machine, can be adapted to solve this problem. The effectiveness and efficiency of the proposed approach have been confirmed by our experiments on three real-world datasets.

Text data is one of the dominating data types in Big Data driven services and applications. The performance of text classification largely depends on the quality of feature extraction over the text corpus. For supervised learning over text documents, the TF-IDF (Term Frequency-Inverse Document Frequency) weighting factor is one of the most frequently used features in text classification. In this paper, we address two known limitations of TF-IDF based feature extraction method: First, the conventional TF-IDF weighting factor lacks of consideration about the synonymous relationship between feature terms. Second, for big corpus with large number of text documents and large number of feature terms, the computational complexity of text classification increases with the dimensionality of the feature space. We address these problems by introducing an optimization technique based on the Inter-Category Distributions (ICD) of terms and the Inter-Category Distributions of documents. We call this new weighting factor TF-IDF-ICD, namely TF-IDF with Inter-Category Distributions. To further enhance the effectiveness of our TF-IDF-ICD method, we describe a TF-IDF-ICD threshold based Dimensionality Reduction (DR) optimization. We test the text classifier with a corpus of 10, 000 articles. The evaluation results show that the proposed TF-IDF-ICD based text classification method outperforms the conventional TF-IDF based classification solution by \(7.84\%\) at only about \(43.19\%\) of the training time used by the conventional TF-IDF based text classification methods.

A novel unsupervised corpus-based stemming technique using lexicon and corpus statistics

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&apos; 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)

All positive examples are alike; each negative example is negative in its own way. During interactive multimedia information retrieval, the number of training samples fed-back by the user is usually small; furthermore, they are not representative for the true distributions-especially the negative examples. Adding to the difficulties is the nonlinearity in real-world distributions. Existing solutions fail to address these problems in a principled way. This paper proposes biased discriminant analysis and transforms specifically designed to address the asymmetry between the positive and negative examples, and to trade off generalization for robustness under a small training sample. The kernel version, namely BiasMap , is derived to facilitate nonlinear biased discrimination. Extensive experiments are carried out for performance evaluation as compared to the state-of-the-art methods.