Discriminative training of the hidden vector state model for semantic parsing

In this paper, we discuss how discriminative training can be applied to the hidden vector state (HVS) model in different task domains. The HVS model is a discrete hidden Markov model (HMM) in which each HMM state represents the state of a push-down automaton with a finite stack size. In previous applications, maximum-likelihood estimation (MLE) is used to derive the parameters of the HVS model. However, MLE makes a number of assumptions and unfortunately some of these assumptions do not hold. Discriminative training, without making such assumptions, can improve the performance of the HVS model by discriminating the correct hypothesis from the competing hypotheses. Experiments have been conducted in two domains: the travel domain for the semantic parsing task using the DARPA Communicator data and the Air Travel Information Services (ATIS) data and the bioinformatics domain for the information extraction task using the GENIA corpus. The results demonstrate modest improvements of the performance of the HVS model using discriminative training. In the travel domain, discriminative training of the HVS model gives a relative error reduction rate of 31 percent in F-measure when compared with MLE on the DARPA Communicator data and 9 percent on the ATIS data. In the bioinformatics domain, a relative error reduction rate of 4 percent in F-measure is achieved on the GENIA corpus.

A hybrid generative/discriminative framework to train a semantic parser from an un-annotated corpus

We propose a hybrid generative/discriminative framework for semantic parsing which combines the hidden vector state (HVS) model and the hidden Markov support vector machines (HM-SVMs). The HVS model is an extension of the basic discrete Markov model in which context is encoded as a stack-oriented state vector. The HM-SVMs combine the advantages of the hidden Markov models and the support vector machines. By employing a modified K-means clustering method, a small set of most representative sentences can be automatically selected from an un-annotated corpus. These sentences together with their abstract annotations are used to train an HVS model which could be subsequently applied on the whole corpus to generate semantic parsing results. The most confident semantic parsing results are selected to generate a fully-annotated corpus which is used to train the HM-SVMs. The proposed framework has been tested on the DARPA Communicator Data. Experimental results show that an improvement over the baseline HVS parser has been observed using the hybrid framework. When compared with the HM-SVMs trained from the fully-annotated corpus, the hybrid framework gave a comparable performance with only a small set of lightly annotated sentences. © 2008. Licensed under the Creative Commons.

Primary and secondary somatosensory cortex responses to anticipation and pain: a magnetoencephalography study

Several brain regions, including the primary and secondary somatosensory cortices (SI and SII, respectively), are functionally active during the pain experience. Both of these regions are thought to be involved in the sensory-discriminative processing of pain and recent evidence suggests that SI in particular may also be involved in more affective processing. In this study we used MEG to investigate the hypothesis that frequency-specific oscillatory activity may be differentially associated with the sensory and affective components of pain. In eight healthy participants (four male), MEG was recorded during a visceral pain experiment comprising baseline, anticipation, pain and post-pain phases. Pain was delivered via intraluminal oesophageal balloon distension (four stimuli at 1 Hz). Significant bilateral but asymmetrical changes in neural activity occurred in the beta-band within SI and SII. In SI, a continuous increase in neural activity occurred during the anticipation phase (20-30 Hz), which continued during the pain phase but at a lower frequency (10-15 Hz). In SII, oscillatory changes only occurred during the pain phase, predominantly in the 20-30 Hz beta band, and were coincident with the stimulus. These data provide novel evidence of functional diversity within SI, indicating a role in attentional and sensory aspects of pain processing. In SII, oscillatory changes were predominantly stimulus-related, indicating a role in encoding the characteristics of the stimulus. We therefore provide objective evidence of functional heterogeneity within SI and functional segregation between SI and SII, and suggest that the temporal and frequency dynamics within cortical regions may offer valuable insights into pain processing.

Predictive gene lists for breast cancer prognosis: a topographic visualisation study

Background: The controversy surrounding the non-uniqueness of predictive gene lists (PGL) of small selected subsets of genes from very large potential candidates as available in DNA microarray experiments is now widely acknowledged 1. Many of these studies have focused on constructing discriminative semi-parametric models and as such are also subject to the issue of random correlations of sparse model selection in high dimensional spaces. In this work we outline a different approach based around an unsupervised patient-specific nonlinear topographic projection in predictive gene lists. Methods: We construct nonlinear topographic projection maps based on inter-patient gene-list relative dissimilarities. The Neuroscale, the Stochastic Neighbor Embedding(SNE) and the Locally Linear Embedding(LLE) techniques have been used to construct two-dimensional projective visualisation plots of 70 dimensional PGLs per patient, classifiers are also constructed to identify the prognosis indicator of each patient using the resulting projections from those visualisation techniques and investigate whether a-posteriori two prognosis groups are separable on the evidence of the gene lists. A literature-proposed predictive gene list for breast cancer is benchmarked against a separate gene list using the above methods. Generalisation ability is investigated by using the mapping capability of Neuroscale to visualise the follow-up study, but based on the projections derived from the original dataset. Results: The results indicate that small subsets of patient-specific PGLs have insufficient prognostic dissimilarity to permit a distinction between two prognosis patients. Uncertainty and diversity across multiple gene expressions prevents unambiguous or even confident patient grouping. Comparative projections across different PGLs provide similar results. Conclusion: The random correlation effect to an arbitrary outcome induced by small subset selection from very high dimensional interrelated gene expression profiles leads to an outcome with associated uncertainty. This continuum and uncertainty precludes any attempts at constructing discriminative classifiers. However a patient's gene expression profile could possibly be used in treatment planning, based on knowledge of other patients' responses. We conclude that many of the patients involved in such medical studies are intrinsically unclassifiable on the basis of provided PGL evidence. This additional category of 'unclassifiable' should be accommodated within medical decision support systems if serious errors and unnecessary adjuvant therapy are to be avoided.

Examination of the predictive value of structural magnetic resonance scans in bipolar disorder:a pattern classification approach

Background - Bipolar disorder (BD) is one of the leading causes of disability worldwide. Patients are further disadvantaged by delays in accurate diagnosis ranging between 5 and 10 years. We applied Gaussian process classifiers (GPCs) to structural magnetic resonance imaging (sMRI) data to evaluate the feasibility of using pattern recognition techniques for the diagnostic classification of patients with BD. Method - GPCs were applied to gray (GM) and white matter (WM) sMRI data derived from two independent samples of patients with BD (cohort 1: n = 26; cohort 2: n = 14). Within each cohort patients were matched on age, sex and IQ to an equal number of healthy controls. Results - The diagnostic accuracy of the GPC for GM was 73% in cohort 1 and 72% in cohort 2; the sensitivity and specificity of the GM classification were respectively 69% and 77% in cohort 1 and 64% and 99% in cohort 2. The diagnostic accuracy of the GPC for WM was 69% in cohort 1 and 78% in cohort 2; the sensitivity and specificity of the WM classification were both 69% in cohort 1 and 71% and 86% respectively in cohort 2. In both samples, GM and WM clusters discriminating between patients and controls were localized within cortical and subcortical structures implicated in BD. Conclusions - Our results demonstrate the predictive value of neuroanatomical data in discriminating patients with BD from healthy individuals. The overlap between discriminative networks and regions implicated in the pathophysiology of BD supports the biological plausibility of the classifiers.

Semi-supervised learning of statistical models for natural language understanding

Natural language understanding is to specify a computational model that maps sentences to their semantic mean representation. In this paper, we propose a novel framework to train the statistical models without using expensive fully annotated data. In particular, the input of our framework is a set of sentences labeled with abstract semantic annotations. These annotations encode the underlying embedded semantic structural relations without explicit word/semantic tag alignment. The proposed framework can automatically induce derivation rules that map sentences to their semantic meaning representations. The learning framework is applied on two statistical models, the conditional random fields (CRFs) and the hidden Markov support vector machines (HM-SVMs). Our experimental results on the DARPA communicator data show that both CRFs and HM-SVMs outperform the baseline approach, previously proposed hidden vector state (HVS) model which is also trained on abstract semantic annotations. In addition, the proposed framework shows superior performance than two other baseline approaches, a hybrid framework combining HVS and HM-SVMs and discriminative training of HVS, with a relative error reduction rate of about 25% and 15% being achieved in F-measure.

The Birmingham relationship continuity measure:the development and evaluation of a measure of the perceived continuity of spousal relationships in dementia

Background: Qualitative research has suggested that spousal carers of someone with dementia differ in terms of whether they perceive their relationship with that person as continuous with the premorbid relationship or as radically different, and that a perception of continuity may be associated with more person-centered care and the experience of fewer of the negative emotions associated with caring. The aim of the study was to develop and evaluate a quantitative measure of the extent to which spousal carers perceive the relationship to be continuous. Methods: An initial pool of 42 questionnaire items was generated on the basis of the qualitative research about relationship continuity. These were completed by 51 spousal carers and item analysis was used to reduce the pool to 23 items. The retained items, comprising five subscales, were then administered to a second sample of 84 spousal carers, and the questionnaire's reliability, discriminative power, and validity were evaluated. Results: The questionnaire showed good reliability: Cronbach's α for the full scale was 0.947, and test-retest reliability was 0.932. Ferguson's δ was 0.987, indicating good discriminative power. Evidence of construct validity was provided by predicted patterns of subscale correlations with the Closeness and Conflict Scale and the Marwit-Meuser Caregiver Grief Inventory. Conclusion: Initial psychometric evaluation of the measure was encouraging. The measure provides a quantitative means of investigating ideas from qualitative research about the role of relationship continuity in influencing how spousal carers provide care and how they react emotionally to their caring role. © 2012 International Psychogeriatric Association.

Privacy and the City:user identification and location semantics in location-based social networks

With the advent of GPS enabled smartphones, an increasing number of users is actively sharing their location through a variety of applications and services. Along with the continuing growth of Location-Based Social Networks (LBSNs), security experts have increasingly warned the public of the dangers of exposing sensitive information such as personal location data. Most importantly, in addition to the geographical coordinates of the user’s location, LBSNs allow easy access to an additional set of characteristics of that location, such as the venue type or popularity. In this paper, we investigate the role of location semantics in the identification of LBSN users. We simulate a scenario in which the attacker’s goal is to reveal the identity of a set of LBSN users by observing their check-in activity. We then propose to answer the following question: what are the types of venues that a malicious user has to monitor to maximize the probability of success? Conversely, when should a user decide whether to make his/her check-in to a location public or not? We perform our study on more than 1 million check-ins distributed over 17 urban regions of the United States. Our analysis shows that different types of venues display different discriminative power in terms of user identity, with most of the venues in the “Residence” category providing the highest re-identification success across the urban regions. Interestingly, we also find that users with a high entropy of their check-ins distribution are not necessarily the hardest to identify, suggesting that it is the collective behaviour of the users’ population that determines the complexity of the identification task, rather than the individual behaviour.

A stable graph-based representation for object recognition through high-order matching

Many Object recognition techniques perform some flavour of point pattern matching between a model and a scene. Such points are usually selected through a feature detection algorithm that is robust to a class of image transformations and a suitable descriptor is computed over them in order to get a reliable matching. Moreover, some approaches take an additional step by casting the correspondence problem into a matching between graphs defined over feature points. The motivation is that the relational model would add more discriminative power, however the overall effectiveness strongly depends on the ability to build a graph that is stable with respect to both changes in the object appearance and spatial distribution of interest points. In fact, widely used graph-based representations, have shown to suffer some limitations, especially with respect to changes in the Euclidean organization of the feature points. In this paper we introduce a technique to build relational structures over corner points that does not depend on the spatial distribution of the features. © 2012 ICPR Org Committee.

Making sense of microposts (#MSM2013) concept extraction challenge

Microposts are small fragments of social media content that have been published using a lightweight paradigm (e.g. Tweets, Facebook likes, foursquare check-ins). Microposts have been used for a variety of applications (e.g., sentiment analysis, opinion mining, trend analysis), by gleaning useful information, often using third-party concept extraction tools. There has been very large uptake of such tools in the last few years, along with the creation and adoption of new methods for concept extraction. However, the evaluation of such efforts has been largely consigned to document corpora (e.g. news articles), questioning the suitability of concept extraction tools and methods for Micropost data. This report describes the Making Sense of Microposts Workshop (#MSM2013) Concept Extraction Challenge, hosted in conjunction with the 2013 World Wide Web conference (WWW'13). The Challenge dataset comprised a manually annotated training corpus of Microposts and an unlabelled test corpus. Participants were set the task of engineering a concept extraction system for a defined set of concepts. Out of a total of 22 complete submissions 13 were accepted for presentation at the workshop; the submissions covered methods ranging from sequence mining algorithms for attribute extraction to part-of-speech tagging for Micropost cleaning and rule-based and discriminative models for token classification. In this report we describe the evaluation process and explain the performance of different approaches in different contexts.

Integrated Omics for the global mapping of biomolecules in LDL

Circulating low density lipoproteins (LDL) are thought to play a crucial role in the onset and development of atherosclerosis, though the detailed molecular mechanisms responsible for their biological effects remain controversial. The complexity of biomolecules (lipids, glycans and protein) and structural features (isoforms and chemical modifications) found in LDL particles hampers the complete understanding of the mechanism underlying its atherogenicity. For this reason the screening of LDL for features discriminative of a particular pathology in search of biomarkers is of high importance. Three major biomolecule classes (lipids, protein and glycans) in LDL particles were screened using mass spectrometry coupled to liquid chromatography. Dual-polarity screening resulted in good lipidome coverage, identifying over 300 lipid species from 12 lipid sub-classes. Multivariate analysis was used to investigate potential discriminators in the individual lipid sub-classes for different study groups (age, gender, pathology). Additionally, the high protein sequence coverage of ApoB-100 routinely achieved (≥70%) assisted in the search for protein modifications correlating to aging and pathology. The large size and complexity of the datasets required the use of chemometric methods (Partial Least Square-Discriminant Analysis, PLS-DA) for their analysis and for the identification of ions that discriminate between study groups. The peptide profile from enzymatically digested ApoB-100 can be correlated with the high structural complexity of lipids associated with ApoB-100 using exploratory data analysis. In addition, using targeted scanning modes, glycosylation sites within neutral and acidic sugar residues in ApoB-100 are also being explored. Together or individually, knowledge of the profiles and modifications of the major biomolecules in LDL particles will contribute towards an in-depth understanding, will help to map the structural features that contribute to the atherogenicity of LDL, and may allow identification of reliable, pathology-specific biomarkers. This research was supported by a Marie Curie Intra-European Fellowship within the 7th European Community Framework Program (IEF 255076). Work of A. Rudnitskaya was supported by Portuguese Science and Technology Foundation, through the European Social Fund (ESF) and "Programa Operacional Potencial Humano - POPH".