Conditional Random Field Classifier Research Articles

Chemical entity recognition in patents by combining dictionary-based and statistical approaches.

We describe the development of a chemical entity recognition system and its application in the CHEMDNER-patent track of BioCreative 2015. This community challenge includes a Chemical Entity Mention in Patents (CEMP) recognition task and a Chemical Passage Detection (CPD) classification task. We addressed both tasks by an ensemble system that combines a dictionary-based approach with a statistical one. For this purpose the performance of several lexical resources was assessed using Peregrine, our open-source indexing engine. We combined our dictionary-based results on the patent corpus with the results of tmChem, a chemical recognizer using a conditional random field classifier. To improve the performance of tmChem, we utilized three additional features, viz. part-of-speech tags, lemmas and word-vector clusters. When evaluated on the training data, our final system obtained an F-score of 85.21% for the CEMP task, and an accuracy of 91.53% for the CPD task. On the test set, the best system ranked sixth among 21 teams for CEMP with an F-score of 86.82%, and second among nine teams for CPD with an accuracy of 94.23%. The differences in performance between the best ensemble system and the statistical system separately were small.Database URL: http://biosemantics.org/chemdner-patents

3D dynamic facial expression recognition using low-resolution videos

In this paper, we focus on the problem of 3D dynamic (4D) facial expression recognition. While traditional methods rely on building deformation models on high-resolution 3D meshes, our approach works directly on low-resolution RGB-D sequences; this feature allows us to apply our algorithm to videos retrieved by widespread and standard low-resolution RGB-D sensors, such as Kinect. After preprocessing both RGB and depth image sequences, sparse features are learned from spatio-temporal local cuboids. Conditional Random Fields classifier is then employed for training and classification. The proposed system is fully-automatic and achieves superior results on three low-resolution datasets built from the 4D facial expression recognition dataset – BU-4DFE. Extensive evaluations of our approach and comparisons with state-of-the-art methods are presented.

The contribution of co-reference resolution to supervised relation detection between bacteria and biotopes entities.

BackgroundThe acquisition of knowledge about relations between bacteria and their locations (habitats and geographical locations) in short texts about bacteria, as defined in the BioNLP-ST 2013 Bacteria Biotope task, depends on the detection of co-reference links between mentions of entities of each of these three types. To our knowledge, no participant in this task has investigated this aspect of the situation. The present work specifically addresses issues raised by this situation: (i) how to detect these co-reference links and associated co-reference chains; (ii) how to use them to prepare positive and negative examples to train a supervised system for the detection of relations between entity mentions; (iii) what context around which entity mentions contributes to relation detection when co-reference chains are provided.ResultsWe present experiments and results obtained both with gold entity mentions (task 2 of BioNLP-ST 2013) and with automatically detected entity mentions (end-to-end system, in task 3 of BioNLP-ST 2013). Our supervised mention detection system uses a linear chain Conditional Random Fields classifier, and our relation detection system relies on a Logistic Regression (aka Maximum Entropy) classifier. They use a set of morphological, morphosyntactic and semantic features. To minimize false inferences, co-reference resolution applies a set of heuristic rules designed to optimize precision. They take into account the types of the detected entity mentions, and take advantage of the didactic nature of the texts of the corpus, where a large proportion of bacteria naming is fairly explicit (although natural referring expressions such as "the bacteria" are common). The resulting system achieved a 0.495 F-measure on the official test set when taking as input the gold entity mentions, and a 0.351 F-measure when taking as input entity mentions predicted by our CRF system, both of which are above the best BioNLP-ST 2013 participant system.ConclusionsWe show that co-reference resolution substantially improves over a baseline system which does not use co-reference information: about 3.5 F-measure points on the test corpus for the end-to-end system (5.5 points on the development corpus) and 7 F-measure points on both development and test corpora when gold mentions are used. While this outperforms the best published system on the BioNLP-ST 2013 Bacteria Biotope dataset, we consider that it provides mostly a stronger baseline from which more work can be started. We also emphasize the importance and difficulty of designing a comprehensive gold standard co-reference annotation, which we explain is a key point to further progress on the task.

Application of acoustic directional data for audio event recognition via HMM/CRF in perimeter surveillance systems

Audio event detection (AED) and recognition is a signal processing and analysis domain used in a wide range of applications including surveillance, home automation and behavioral assessment. The field presents numerous challenges to the current state-of-the-art due to its highly nonlinear nature. High false alarm rates (FARs) in such applications particularly limit the capabilities of vision-based perimeter monitoring systems by inducing high operator dependence. On the other hand, conventional fence-based vibration detectors and pressure-driven “taut wires” offer high sensitivity at the cost of a high FAR due to debris, animals and weather.This work reports an audio event identification methodology implemented as a test-bed system for a surveillance application to reduce FAR, maximize blind-spot coverage and improve audio event classification accuracy. The first phase utilizes a nonlinear autoregressive classifier to locate and classify discrete audio events via an exogenous sound direction variable to improve classifier confidence. The second phase implements a time-series-based system to recognize various audio activity groups from nominal everyday sound events such as traffic and muffled speech. The discretely labeled data is thus trained with HMM and Conditional Random Field classifiers and reports a substantial improvement in classification accuracies of indoor human activities.

Improving chemical entity recognition through h-index based semantic similarity.

BackgroundOur approach to the BioCreative IV challenge of recognition and classification of drug names (CHEMDNER task) aimed at achieving high levels of precision by applying semantic similarity validation techniques to Chemical Entities of Biological Interest (ChEBI) mappings. Our assumption is that the chemical entities mentioned in the same fragment of text should share some semantic relation. This validation method was further improved by adapting the semantic similarity measure to take into account the h-index of each ancestor. We applied this method in two measures, simUI and simGIC, and validated the results obtained for the competition, comparing each adapted measure to its original version.ResultsFor the competition, we trained a Random Forest classifier that uses various scores provided by our system, including semantic similarity, which improved the F-measure obtained with the Conditional Random Fields classifiers by 4.6%. Using a notion of concept relevance based on the h-index measure, we were able to enhance our validation process so that for a fixed recall, we increased precision by excluding from the results a higher amount of false positives. We plotted precision and recall values for a range of validation thresholds using different similarity measures, obtaining higher precision values for the same recall with the measures based on the h-index.ConclusionsThe semantic similarity measure we introduced was more efficient at validating text mining results from machine learning classifiers than other measures. We improved the results we obtained for the CHEMDNER task by maintaining high precision values while improving the recall and F-measure.

A Support Vector Conditional Random Fields Classifier With a Mahalanobis Distance Boundary Constraint for High Spatial Resolution Remote Sensing Imagery

In this paper, a modified conditional random fields (CRFs) classifier, namely the support vector conditional random fields classifier with a Mahalanobis distance boundary constraint (SVRFMC), is proposed to perform the task of classification for high spatial resolution (HSR) remote sensing imagery. In SVRFMC, the CRFs model has the intrinsic ability of incorporating the contextual information in both the observation and labeling fields. Support vector machine (SVM) is set as the spectral term to get a more precise estimation of each pixel's probability of belonging to each possible class. To preserve the spatial details in the classification result, a Mahalanobis distance boundary constraint is considered as the spatial term to undertake appropriate spatial smoothing. By integrating SVM and a Mahalanobis distance boundary constraint, SVRFMC can not only avoid the explicit modeling of observed data, but can also undertake appropriate smoothing with the consideration of contextual information, thereby exhibiting more universality and validity in the application of HSR image classification, especially when the image has a complex land-cover class distribution and the training samples are limited. Three HSR images comprising QuickBird, IKONOS, and HYDICE imagery were utilized to evaluate the performance of the proposed algorithm in comparison to other image classification approaches: noncontextual multiclass SVM, a traditional object-oriented classifier (OOC), an object-oriented classification based on fractal net evolution approach (FNEA) segmentation (OO-FNEA), a simplified CRF model with boundary constraint (BC-CRF), and a recently proposed contextual classifier combining SVM and Markov random fields (Markovian support vector classifier). The experimental results demonstrate that the SVRFMC algorithm is superior to the other methods, providing a satisfactory classification result for HSR imagery, including both multispectral HSR imagery and hyperspectral HSR imagery, even with limited training samples, from both the visualization and quantitative evaluations.

Towards a Protein–Protein Interaction information extraction system: Recognizing named entities

The majority of biological functions of any living being are related to Protein–Protein Interactions (PPI). PPI discoveries are reported in form of research publications whose volume grows day after day. Consequently, automatic PPI information extraction systems are a pressing need for biologists. In this paper we are mainly concerned with the named entity detection module of PPIES (the PPI information extraction system we are implementing) which recognizes twelve entity types relevant in PPI context. It is composed of two sub-modules: a dictionary look-up with extensive normalization and acronym detection, and a Conditional Random Field classifier. The dictionary look-up module has been tested with Interaction Method Task (IMT), and it improves by approximately 10% the current solutions that do not use Machine Learning (ML). The second module has been used to create a classifier using the Joint Workshop on Natural Language Processing in Biomedicine and its Applications (JNLPBA’04) data set. It does not use any external resources, or complex or ad hoc post-processing, and obtains 77.25%, 75.04% and 76.13 for precision, recall, and F1-measure, respectively, improving all previous results obtained for this data set.

Towards subject independent continuous sign language recognition: A segment and merge approach

This paper presents a segment-based probabilistic approach to robustly recognize continuous sign language sentences. The recognition strategy is based on a two-layer conditional random field (CRF) model, where the lower layer processes the component channels and provides outputs to the upper layer for sign recognition. The continuously signed sentences are first segmented, and the sub-segments are labeled SIGN or ME (movement epenthesis) by a Bayesian network (BN) which fuses the outputs of independent CRF and support vector machine (SVM) classifiers. The sub-segments labeled as ME are discarded and the remaining SIGN sub-segments are merged and recognized by the two-layer CRF classifier; for this we have proposed a new algorithm based on the semi-Markov CRF decoding scheme. With eight signers, we obtained a recall rate of 95.7% and a precision of 96.6% for unseen samples from seen signers, and a recall rate of 86.6% and a precision of 89.9% for unseen signers.

Scalable sequential alternating proximal methods for sparse structural SVMs and CRFs

Structural Support Vector Machines (SSVMs) and Conditional Random Fields (CRFs) are popular discriminative methods used for classifying structured and complex objects like parse trees, image segments and part-of-speech tags. The datasets involved are very large dimensional, and the models designed using typical training algorithms for SSVMs and CRFs are non-sparse. This non-sparse nature of models results in slow inference. Thus, there is a need to devise new algorithms for sparse SSVM and CRF classifier design. Use of elastic net and L1-regularizer has already been explored for solving primal CRF and SSVM problems, respectively, to design sparse classifiers. In this work, we focus on dual elastic net regularized SSVM and CRF. By exploiting the weakly coupled structure of these convex programming problems, we propose a new sequential alternating proximal (SAP) algorithm to solve these dual problems. This algorithm works by sequentially visiting each training set example and solving a simple subproblem restricted to a small subset of variables associated with that example. Numerical experiments on various benchmark sequence labeling datasets demonstrate that the proposed algorithm scales well. Further, the classifiers designed are sparser than those designed by solving the respective primal problems and demonstrate comparable generalization performance. Thus, the proposed SAP algorithm is a useful alternative for sparse SSVM and CRF classifier design.

Comprehensive temporal information detection from clinical text: medical events, time, and TLINK identification

Temporal information detection systems have been developed by the Mayo Clinic for the 2012 i2b2 Natural Language Processing Challenge. To construct automated systems for EVENT/TIMEX3 extraction and temporal link (TLINK) identification from clinical text. The i2b2 organizers provided 190 annotated discharge summaries as the training set and 120 discharge summaries as the test set. Our Event system used a conditional random field classifier with a variety of features including lexical information, natural language elements, and medical ontology. The TIMEX3 system employed a rule-based method using regular expression pattern match and systematic reasoning to determine normalized values. The TLINK system employed both rule-based reasoning and machine learning. All three systems were built in an Apache Unstructured Information Management Architecture framework. Our TIMEX3 system performed the best (F-measure of 0.900, value accuracy 0.731) among the challenge teams. The Event system produced an F-measure of 0.870, and the TLINK system an F-measure of 0.537. Our TIMEX3 system demonstrated good capability of regular expression rules to extract and normalize time information. Event and TLINK machine learning systems required well-defined feature sets to perform well. We could also leverage expert knowledge as part of the machine learning features to further improve TLINK identification performance.

Mining FDA drug labels for medical conditions

BackgroundCincinnati Children’s Hospital Medical Center (CCHMC) has built the initial Natural Language Processing (NLP) component to extract medications with their corresponding medical conditions (Indications, Contraindications, Overdosage, and Adverse Reactions) as triples of medication-related information ([(1) drug name]-[(2) medical condition]-[(3) LOINC section header]) for an intelligent database system, in order to improve patient safety and the quality of health care. The Food and Drug Administration’s (FDA) drug labels are used to demonstrate the feasibility of building the triples as an intelligent database system task.MethodsThis paper discusses a hybrid NLP system, called AutoMCExtractor, to collect medical conditions (including disease/disorder and sign/symptom) from drug labels published by the FDA. Altogether, 6,611 medical conditions in a manually-annotated gold standard were used for the system evaluation. The pre-processing step extracted the plain text from XML file and detected eight related LOINC sections (e.g. Adverse Reactions, Warnings and Precautions) for medical condition extraction. Conditional Random Fields (CRF) classifiers, trained on token, linguistic, and semantic features, were then used for medical condition extraction. Lastly, dictionary-based post-processing corrected boundary-detection errors of the CRF step. We evaluated the AutoMCExtractor on manually-annotated FDA drug labels and report the results on both token and span levels.ResultsPrecision, recall, and F-measure were 0.90, 0.81, and 0.85, respectively, for the span level exact match; for the token-level evaluation, precision, recall, and F-measure were 0.92, 0.73, and 0.82, respectively.ConclusionsThe results demonstrate that (1) medical conditions can be extracted from FDA drug labels with high performance; and (2) it is feasible to develop a framework for an intelligent database system.

Stacked ensemble coupled with feature selection for biomedical entity extraction

Entity extraction is one of the most fundamental and important tasks in biomedical information extraction. In this paper we propose a two-stage algorithm for the extraction of biomedical entities in the forms of genes and gene product mentions in text. Several different approaches have emerged but most of these state-of-the-art approaches suggest that individual system may not cover entity representations with arbitrary set of features and cannot achieve best performance. We identify and implement a diverse set of features which are relevant for the identification of biomedical entities and classification of them into some predefined categories. One most important criterion of these features is that these are identified and selected largely without using any domain knowledge. In the first stage we use a genetic algorithm (GA) based feature selection technique to determine the most relevant set of features for Support Vector Machine (SVM) and Conditional Random Field (CRF) classifiers. The GA based feature selection algorithm produces best population that can be used to generate different classification models based on CRF and SVM. In the second stage we develop a stacked based ensemble to combine the classifiers selected in the first stage. The proposed approach is evaluated on two benchmark datasets, namely JNLPBA 2004 shared task and GENETAG. The proposed approach yields the overall F-measure values of 75.17% and 94.70% for JNLPBA 2004 and GENETAG data sets, respectively.

Combining an Expert-Based Medical Entity Recognizer to a Machine-Learning System: Methods and a Case Study

Medical entity recognition is currently generally performed by data-driven methods based on supervised machine learning. Expert-based systems, where linguistic and domain expertise are directly provided to the system are often combined with data-driven systems. We present here a case study where an existing expert-based medical entity recognition system, Ogmios, is combined with a data-driven system, Caramba, based on a linear-chain Conditional Random Field (CRF) classifier. Our case study specifically highlights the risk of overfitting incurred by an expert-based system. We observe that it prevents the combination of the 2 systems from obtaining improvements in precision, recall, or F-measure, and analyze the underlying mechanisms through a post-hoc feature-level analysis. Wrapping the expert-based system alone as attributes input to a CRF classifier does boost its F-measure from 0.603 to 0.710, bringing it on par with the data-driven system. The generalization of this method remains to be further investigated.

Exploring the Trade-off Between Accuracy and Observational Latency in Action Recognition

An important aspect in designing interactive, action-based interfaces is reliably recognizing actions with minimal latency. High latency causes the system's feedback to lag behind user actions and thus significantly degrades the interactivity of the user experience. This paper presents algorithms for reducing latency when recognizing actions. We use a latency-aware learning formulation to train a logistic regression-based classifier that automatically determines distinctive canonical poses from data and uses these to robustly recognize actions in the presence of ambiguous poses. We introduce a novel (publicly released) dataset for the purpose of our experiments. Comparisons of our method against both a Bag of Words and a Conditional Random Field (CRF) classifier show improved recognition performance for both pre-segmented and online classification tasks. Additionally, we employ GentleBoost to reduce our feature set and further improve our results. We then present experiments that explore the accuracy/latency trade-off over a varying number of actions. Finally, we evaluate our algorithm on two existing datasets.

Weighted Conditional Random Fields for Supervised Interpatient Heartbeat Classification

This paper proposes a method for the automatic classification of heartbeats in an ECG signal. Since this task has specific characteristics such as time dependences between observations and a strong class unbalance, a specific classifier is proposed and evaluated on real ECG signals from the MIT arrhythmia database. This classifier is a weighted variant of the conditional random fields classifier. Experiments show that the proposed method outperforms previously reported heartbeat classification methods, especially for the pathological heartbeats.

A Semi-Markov Model for Mitosis Segmentation in Time-Lapse Phase Contrast Microscopy Image Sequences of Stem Cell Populations

We propose a semi-Markov model trained in a max-margin learning framework for mitosis event segmentation in large-scale time-lapse phase contrast microscopy image sequences of stem cell populations. Our method consists of three steps. First, we apply a constrained optimization based microscopy image segmentation method that exploits phase contrast optics to extract candidate subsequences in the input image sequence that contains mitosis events. Then, we apply a max-margin hidden conditional random field (MM-HCRF) classifier learned from human-annotated mitotic and nonmitotic sequences to classify each candidate subsequence as a mitosis or not. Finally, a max-margin semi-Markov model (MM-SMM) trained on manually-segmented mitotic sequences is utilized to reinforce the mitosis classification results, and to further segment each mitosis into four predefined temporal stages. The proposed method outperforms the event-detection CRF model recently reported by Huh as well as several other competing methods in very challenging image sequences of multipolar-shaped C3H10T1/2 mesenchymal stem cells. For mitosis detection, an overall precision of 95.8% and a recall of 88.1% were achieved. For mitosis segmentation, the mean and standard deviation for the localization errors of the start and end points of all mitosis stages were well below 1 and 2 frames, respectively. In particular, an overall temporal location error of 0.73 ± 1.29 frames was achieved for locating daughter cell birth events.

EVENT AND TEMPORAL EXPRESSION EXTRACTION FROM RAW TEXT: FIRST STEP TOWARDS A TEMPORALLY AWARE SYSTEM

Extracting temporal information from raw text is fundamental for deep language understanding, and key to many applications like question answering, information extraction, and document summarization. Our long-term goal is to build complete temporal structure of documents and use the temporal structure in other applications like textual entailment, question answering, visualization, or others. In this paper, we present a first step, a system for extracting events, event features, main events, temporal expressions and their normalized values from raw text. Our system is a combination of deep semantic parsing with extraction rules, Markov Logic Network classifiers and Conditional Random Field classifiers. To compare with existing systems, we evaluated our system on the TempEval-1 and TempEval-2 corpus. Our system outperforms or performs competitively with existing systems that evaluate on the TimeBank, TempEval-1 and TempEval-2 corpus and our performance is very close to inter-annotator agreement of the TimeBank annotators.

Sentiment Analysis in Weblog Using Contextual Information: A Machine Learning Approach

This paper investigates sentiments in blog data using machine learning techniques such as support vector machines (SVM) and conditional random fields (CRF). Bloggers collaboratively contribute to the formation of blog corpora, which contain a rich set of icon-tagged text expressions. First, a collocation model on the expressions is used to construct a lexicon which provides keyword features for subsequent classification models. SVM and CRF classifiers use these keywords as features to assign a sentence sentiment bipolarity or arousal-valence space categories. Our methods determine a sentiment category by taking contexts into account. Experiments show that CRF classifiers outperform SVM classifiers and the performance benefits from increasing the training size of blog materials.