Positive Training Examples Research Articles

Public Health Security Systems Empowered by Artificial Intelligence for Early Monitoring and Prevention of Epidemics

Artificial intelligence (AI) methods have been extensively used for detecting and predicting Infectious Disease (ID) outbreaks as time series and modeling and evaluating Public Health responses. The significant tasks of PH monitoring and intervention present distinct technical difficulties, including limited data availability, absence of sufficient positive training examples, challenges in establishing benchmarks, measuring the effectiveness of management policies, complex relationships between spatial and time series elements, and more detailed risk assessments involving interaction and social networks. Conventional PH monitoring mainly depends on statistical methods. In recent years, there has been a significant expansion of approaches that AI enables. This research presents an AI method called Early Monitoring and Prevention of Epidemics (AI-EMPE) for enhancing PH security systems. The suggested approach converts a substantial amount of collected PH security incidents into separate incident characteristics and utilizes a Deep Learning (DL)--based detection technique to enhance EMPE. AI-EMPE incorporates sophisticated AI methods, including integrated Convolutional Neural Networks (CNN) and Backpropagation Neural Networks (BP-NN). These findings indicate that the integrated method is the most efficient in improving PH security systems.

Efficiency of machine learning to predict gas accumulations

A growing number of research geophysicists are drawn to machine learning in geophysical applications. At the moment, direct applications of machine learning cover an expansive range of tasks related to the interpretation of seismic data. Beginning with computer-assisted seismic well tying, faults and horizons are automatically interpreted, velocity models are optimized, geobodies are extracted, and seismic facies are classified. Our goal is to demonstrate how artificial neural networks may be used to teach a computer to categorize seismic data and anticipate possible gas accumulations, starting from selecting training data, through filtering the best discriminating attributes, monitoring the learning process, and ending with running the predictions. This workflow was applied on a Dana Gas Messininan gas field in the Nile Delta province of Egypt. The network was trained on a few seismic points in the vicinity of one successful well as positive training examples. The network not only accurately predicted the gas accumulation near the training points but also the other nearby stratigraphically compartmentalized gas accumulations, aligning with production and pressure data of the hidden wells. Similar findings were obtained when the results were compared to classical amplitude variation with offset elastic inversion; however, the machine learning methodology is far faster and easier than the inversion in terms of procedural steps and required work.

Automated identification of clinical features from sparsely annotated 3-dimensional medical imaging

One of the core challenges in applying machine learning and artificial intelligence to medicine is the limited availability of annotated medical data. Unlike in other applications of machine learning, where an abundance of labeled data is available, the labeling and annotation of medical data and images require a major effort of manual work by expert clinicians who do not have the time to annotate manually. In this work, we propose a new deep learning technique (SLIVER-net), to predict clinical features from 3-dimensional volumes using a limited number of manually annotated examples. SLIVER-net is based on transfer learning, where we borrow information about the structure and parameters of the network from publicly available large datasets. Since public volume data are scarce, we use 2D images and account for the 3-dimensional structure using a novel deep learning method which tiles the volume scans, and then adds layers that leverage the 3D structure. In order to illustrate its utility, we apply SLIVER-net to predict risk factors for progression of age-related macular degeneration (AMD), a leading cause of blindness, from optical coherence tomography (OCT) volumes acquired from multiple sites. SLIVER-net successfully predicts these factors despite being trained with a relatively small number of annotated volumes (hundreds) and only dozens of positive training examples. Our empirical evaluation demonstrates that SLIVER-net significantly outperforms standard state-of-the-art deep learning techniques used for medical volumes, and its performance is generalizable as it was validated on an external testing set. In a direct comparison with a clinician panel, we find that SLIVER-net also outperforms junior specialists, and identifies AMD progression risk factors similarly to expert retina specialists.

Hierarchical Adaptive Multi-task Learning Framework for Patient Diagnoses and Diagnostic Category Classification.

The problems a patient suffers from can be summarized in terms of a list of patient diagnoses. The diagnoses are typically organized in a hierarchy (or a lattice structure) in which many different low-level diagnoses are covered by one or more diagnostic categories. An interesting machine learning problem is related to learning of a wide range of diagnostic models (at different levels of abstraction) that can automatically assign a diagnosis or a diagnostic category to a specific patient. While one can always approach this problem by learning models for each diagnostic task independently, an interesting open question is how one can leverage the knowledge of a diagnostic hierarchy to improve the classification and outperform independent diagnostic models. In this work, we study this problem by designing a new hierarchical classification learning framework in which multiple diagnostic classification targets are explicitly related via diagnostic hierarchy relations. By conducting experiments on MIMIC-III data and ICD-9 diagnosis hierarchy, we demonstrate that our framework leads to improved classification performance on individual diagnostic tasks when compared to independently learned diagnostic models. This improvement is stronger for diagnoses with a low prior and smaller number of positive training examples.

Adversarial Learning of Semantic Relevance in Text to Image Synthesis

We describe a new approach that improves the training of generative adversarial nets (GANs) for synthesizing diverse images from a text input. Our approach is based on the conditional version of GANs and expands on previous work leveraging an auxiliary task in the discriminator. Our generated images are not limited to certain classes and do not suffer from mode collapse while semantically matching the text input. A key to our training methods is how to form positive and negative training examples with respect to the class label of a given image. Instead of selecting random training examples, we perform negative sampling based on the semantic distance from a positive example in the class. We evaluate our approach using the Oxford-102 flower dataset, adopting the inception score and multi-scale structural similarity index (MS-SSIM) metrics to assess discriminability and diversity of the generated images. The empirical results indicate greater diversity in the generated images, especially when we gradually select more negative training examples closer to a positive example in the semantic space.

Design of a generic, open platform for machine learning-assisted indexing and clustering of articles in PubMed, a biomedical bibliographic database

AbstractMany investigators have carried out text mining of the biomedical literature for a variety of purposes, ranging from the assignment of indexing terms to the disambiguation of author names. A common approach is to define positive and negative training examples, extract features from article metadata, and use machine learning algorithms. At present, each research group tackles each problem from scratch, in isolation of other projects, which causes redundancy and a great waste of effort. Here, we propose and describe the design of a generic platform for biomedical text mining, which can serve as a shared resource for machine learning projects and as a public repository for their outputs. We initially focus on a specific goal, namely, classifying articles according to publication type and emphasize how feature sets can be made more powerful and robust through the use of multiple, heterogeneous similarity measures as input to machine learning models. We then discuss how the generic platform can be extended to include a wide variety of other machine learning-based goals and projects and can be used as a public platform for disseminating the results of natural language processing (NLP) tools to end-users as well.

Rapid broad area search and detection of Chinese surface-to-air missile sites using deep convolutional neural networks

We evaluated how deep convolutional neural networks (DCNN) could assist in the labor-intensive process of human visual searches for objects of interest in high-resolution imagery over large areas of the Earth’s surface. Various DCNN were trained and tested using fewer than 100 positive training examples (China only) from a worldwide surface-to-air-missile (SAM) site dataset. A ResNet-101 DCNN achieved a 98.2% average accuracy for the China SAM site data. The ResNet-101 DCNN was used to process ∼19.6 M image chips over a large study area in southeastern China. DCNN chip detections (∼9300) were postprocessed with a spatial clustering algorithm to produce a ranked list of ∼2100 candidate SAM site locations. The combination of DCNN processing and spatial clustering effectively reduced the search area by ∼660X (0.15% of the DCNN-processed land area). An efficient web interface was used to facilitate a rapid serial human review of the candidate SAM sites in the China study area. Four novice imagery analysts with no prior imagery analysis experience were able to complete a DCNN-assisted SAM site search in an average time of ∼42 min. This search was ∼81X faster than a traditional visual search over an equivalent land area of ∼88,640 km2 while achieving nearly identical statistical accuracy (∼90% F1).

Exploiting Unlabeled Data for Neural Grammatical Error Detection

Identifying and correcting grammatical errors in the text written by non-native writers have received increasing attention in recent years. Although a number of annotated corpora have been established to facilitate data-driven grammatical error detection and correction approaches, they are still limited in terms of quantity and coverage because human annotation is labor-intensive, time-consuming, and expensive. In this work, we propose to utilize unlabeled data to train neural network based grammatical error detection models. The basic idea is to cast error detection as a binary classification problem and derive positive and negative training examples from unlabeled data. We introduce an attention-based neural network to capture long-distance dependencies that influence the word being detected. Experiments show that the proposed approach significantly outperforms SVM and convolutional networks with fixed-size context window.

Influence of Varying Training Set Composition and Size on Support Vector Machine-Based Prediction of Active Compounds

Support vector machine (SVM) modeling is one of the most popular machine learning approaches in chemoinformatics and drug design. The influence of training set composition and size on predictions currently is an underinvestigated issue in SVM modeling. In this study, we have derived SVM classification and ranking models for a variety of compound activity classes under systematic variation of the number of positive and negative training examples. With increasing numbers of negative training compounds, SVM classification calculations became increasingly accurate and stable. However, this was only the case if a required threshold of positive training examples was also reached. In addition, consideration of class weights and optimization of cost factors substantially aided in balancing the calculations for increasing numbers of negative training examples. Taken together, the results of our analysis have practical implications for SVM learning and the prediction of active compounds. For all compound classes under study, top recall performance and independence of compound recall of training set composition was achieved when 250–500 active and 500–1000 randomly selected inactive training instances were used. However, as long as ∼50 known active compounds were available for training, increasing numbers of 500–1000 randomly selected negative training examples significantly improved model performance and gave very similar results for different training sets.

Learning and Calibrating Per-Location Classifiers for Visual Place Recognition

The aim of this work is to localize a query photograph by finding other images depicting the same place in a large geotagged image database. This is a challenging task due to changes in viewpoint, imaging conditions and the large size of the image database. The contribution of this work is two-fold. First, we cast the place recognition problem as a classification task and use the available geotags to train a classifier for each location in the database in a similar manner to per-exemplar SVMs in object recognition. Second, as only one or a few positive training examples are available for each location, we propose two methods to calibrate all the per-location SVM classifiers without the need for additional positive training data. The first method relies on p-values from statistical hypothesis testing and uses only the available negative training data. The second method performs an affine calibration by appropriately normalizing the learnt classifier hyperplane and does not need any additional labelled training data. We test the proposed place recognition method with the bag-of-visual-words and Fisher vector image representations suitable for large scale indexing. Experiments are performed on three datasets: 25,000 and 55,000 geotagged street view images of Pittsburgh, and the 24/7 Tokyo benchmark containing 76,000 images with varying illumination conditions. The results show improved place recognition accuracy of the learnt image representation over direct matching of raw image descriptors.

Feature Selection Has a Large Impact on One-Class Classification Accuracy for MicroRNAs in Plants.

MicroRNAs (miRNAs) are short RNA sequences involved in posttranscriptional gene regulation. Their experimental analysis is complicated and, therefore, needs to be supplemented with computational miRNA detection. Currently computational miRNA detection is mainly performed using machine learning and in particular two-class classification. For machine learning, the miRNAs need to be parametrized and more than 700 features have been described. Positive training examples for machine learning are readily available, but negative data is hard to come by. Therefore, it seems prerogative to use one-class classification instead of two-class classification. Previously, we were able to almost reach two-class classification accuracy using one-class classifiers. In this work, we employ feature selection procedures in conjunction with one-class classification and show that there is up to 36% difference in accuracy among these feature selection methods. The best feature set allowed the training of a one-class classifier which achieved an average accuracy of ~95.6% thereby outperforming previous two-class-based plant miRNA detection approaches by about 0.5%. We believe that this can be improved upon in the future by rigorous filtering of the positive training examples and by improving current feature clustering algorithms to better target pre-miRNA feature selection.

Do We Need More Training Data?

Datasets for training object recognition systems are steadily increasing in size. This paper investigates the question of whether existing detectors will continue to improve as data grows, or saturate in performance due to limited model complexity and the Bayes risk associated with the feature spaces in which they operate. We focus on the popular paradigm of discriminatively trained templates defined on oriented gradient features. We investigate the performance of mixtures of templates as the number of mixture components and the amount of training data grows. Surprisingly, even with proper treatment of regularization and “outliers”, the performance of classic mixture models appears to saturate quickly ( $${\sim }10$$ templates and $${\sim }100$$ positive training examples per template). This is not a limitation of the feature space as compositional mixtures that share template parameters via parts and that can synthesize new templates not encountered during training yield significantly better performance. Based on our analysis, we conjecture that the greatest gains in detection performance will continue to derive from improved representations and learning algorithms that can make efficient use of large datasets.

Clustering-based Method for Positive and Unlabeled Text Categorization Enhanced by Improved TFIDF

PU learning occurs frequently in Web pages classification and text retrieval applications because users may be interested in information on the same topic. Collecting reliable negative examples is a key step in PU (Positive and Unlabeled) text classification, which solves a key problem in machine learning when no labeled negative examples are available in the training set or negative examples are difficult to collect. Thus, this paper presents a novel clustering-based method for collecting reliable negative examples (CCRNE). Different from traditional methods, we remove as many probable positive examples from unlabeled set as possible, which results that more reliable negative examples are found out. During the process of building classifier, a novel TFIDF-improved feature weighting approach, which reflects the importance of the term in the positive and negative training examples respectively, is presented to describe documents in the Vector Space Model. We also build a weighted voting classifier by iteratively applying the SVM algorithm and implement OCS (One-class SVM), PEBL (Positive Example Based Learning) and 1-DNFII (Constrained 1-DNF) methods used for comparison. Experimental results on three real-world datasets (Reuters Corpus Volume 1 (RCV1), Reuters-21578 and 20 Newsgroups) show that our proposed C-CRNE extracts more reliable negative examples than the baseline algorithms with very low error rates. And our classifier outperforms other state-of-art classification methods from the perspective of traditional performance metrics.

Best practices for learning video concept detectors from social media examples

Learning video concept detectors from social media sources, such as Flickr images and YouTube videos, has the potential to address a wide variety of concept queries for video search. While the potential has been recognized by many, and progress on the topic has been impressive, we argue that key questions crucial to know how to learn effective video concept detectors from social media examples? remain open. As an initial attempt to answer these questions, we conduct an experimental study using a video search engine which is capable of learning concept detectors from social media examples, be it socially tagged videos or socially tagged images. Within the video search engine we investigate three strategies for positive example selection, three negative example selection strategies and three learning strategies. The performance is evaluated on the challenging TRECVID 2012 benchmark consisting of 600 h of Internet video. From the experiments we derive four best practices: (1) tagged images are a better source for learning video concepts than tagged videos, (2) selecting tag relevant positive training examples is always beneficial, (3) selecting relevant negative examples is advantageous and should be treated differently for video and image sources, and (4) learning concept detectors with selected relevant training data before learning is better then incorporating the relevance during the learning process. The best practices within our video search engine lead to state-of-the-art performance in the TRECVID 2013 benchmark for concept detection without manually provided annotations.

Discriminatively Reranking Abductive Proofs for Plan Recognition

We investigate the use of a simple, discriminative reranking approach to plan recognition in an abductive setting. In contrast to recent work, which attempts to model abductive plan recognition using various formalisms that integrate logic and graphical models (such as Markov Logic Networks or Bayesian Logic Programs), we instead advocate a simpler, more flexible approach in which plans found through an abductive beam-search are discriminatively scored based on arbitrary features. We show that this approach performs well even with relatively few positive training examples, and we obtain state-of-the-art results on two abductive plan recognition datasets, outperforming more complicated systems.

Multimedia Event Detection Using A Classifier-Specific Intermediate Representation

Multimedia event detection (MED) plays an important role in many applications such as video indexing and retrieval. Current event detection works mainly focus on sports and news event detection or abnormality detection in surveillance videos. Differently, our research aims to detect more complicated and generic events within a longer video sequence. In the past, researchers have proposed using intermediate concept classifiers with concept lexica to help understand the videos. Yet it is difficult to judge how many and what concepts would be sufficient for the particular video analysis task. Additionally, obtaining robust semantic concept classifiers requires a large number of positive training examples, which in turn has high human annotation cost. In this paper, we propose an approach that exploits the external concepts-based videos and event-based videos simultaneously to learn an intermediate representation from video features. Our algorithm integrates the classifier inference and latent intermediate representation into a joint framework. The joint optimization of the intermediate representation and the classifier makes them mutually beneficial and reciprocal. Effectively, the intermediate representation and the classifier are tightly correlated. The classifier dependent intermediate representation not only accurately reflects the task semantics but is also more suitable for the specific classifier. Thus we have created a discriminative semantic analysis framework based on a tightly coupled intermediate representation. Extensive experiments on multimedia event detection using real-world videos demonstrate the effectiveness of the proposed approach.

Can MiRBase Provide Positive Data for Machine Learning for the Detection of MiRNA Hairpins?

Summary Experimental detection and validation of miRNAs is a tedious, time-consuming, and expensive process. Computational methods for miRNA gene detection are being developed so that the number of candidates that need experimental validation can be reduced to a manageable amount. Computational methods involve homology-based and ab inito algorithms. Both approaches are dependent on positive and negative training examples. Positive examples are usually derived from miRBase, the main resource for experimentally validated miRNAs. We encountered some problems with miRBase which we would like to report here. Some problems, among others, we encountered are that folds presented in miRBase are not always the fold with the minimum free energy; some entries do not seem to conform to expectations of miRNAs, and some external accession numbers are not valid. In addition, we compared the prediction accuracy for the same negative dataset when the positive data came from miRBase or miRTarBase and found that the latter led to more precise prediction models. We suggest that miRBase should introduce some automated facilities for ensuring data quality to overcome these problems.

Can MiRBase provide positive data for machine learning for the detection of MiRNA hairpins?

Experimental detection and validation of miRNAs is a tedious, time-consuming, and expensive process. Computational methods for miRNA gene detection are being developed so that the number of candidates that need experimental validation can be reduced to a manageable amount. Computational methods involve homology-based and ab inito algorithms. Both approaches are dependent on positive and negative training examples. Positive examples are usually derived from miRBase, the main resource for experimentally validated miRNAs. We encountered some problems with miRBase which we would like to report here. Some problems, among others, we encountered are that folds presented in miRBase are not always the fold with the minimum free energy; some entries do not seem to conform to expectations of miRNAs, and some external accession numbers are not valid. In addition, we compared the prediction accuracy for the same negative dataset when the positive data came from miRBase or miRTarBase and found that the latter led to more precise prediction models. We suggest that miRBase should introduce some automated facilities for ensuring data quality to overcome these problems.

Flaws in evaluation schemes for pair-input computational predictions

Flaws in evaluation schemes for pair-input computational predictions

Machine learning with naturally labeled data for identifying abbreviation definitions

BackgroundThe rapid growth of biomedical literature requires accurate text analysis and text processing tools. Detecting abbreviations and identifying their definitions is an important component of such tools. Most existing approaches for the abbreviation definition identification task employ rule-based methods. While achieving high precision, rule-based methods are limited to the rules defined and fail to capture many uncommon definition patterns. Supervised learning techniques, which offer more flexibility in detecting abbreviation definitions, have also been applied to the problem. However, they require manually labeled training data.MethodsIn this work, we develop a machine learning algorithm for abbreviation definition identification in text which makes use of what we term naturally labeled data. Positive training examples are naturally occurring potential abbreviation-definition pairs in text. Negative training examples are generated by randomly mixing potential abbreviations with unrelated potential definitions. The machine learner is trained to distinguish between these two sets of examples. Then, the learned feature weights are used to identify the abbreviation full form. This approach does not require manually labeled training data.ResultsWe evaluate the performance of our algorithm on the Ab3P, BIOADI and Medstract corpora. Our system demonstrated results that compare favourably to the existing Ab3P and BIOADI systems. We achieve an F-measure of 91.36% on Ab3P corpus, and an F-measure of 87.13% on BIOADI corpus which are superior to the results reported by Ab3P and BIOADI systems. Moreover, we outperform these systems in terms of recall, which is one of our goals.