Context Of Annotation Research Articles

878-P: Beyond Tracking: The Benefits of Contextual Annotation in a Diabetes Digital Therapeutic

Objective: Digital therapeutics manage chronic diseases by leveraging structured data like blood glucose/pressure, diet, etc. Addition of free-text notes associated with this data may help patients/providers make sense of trends by tracking unique concerns, providing the benefits of deeper cognitive processing. We explored the impact of contextual annotation in a digital therapeutic on A1C. Method and Results: Free-text note usage from 3,142 users of BlueStar® digital therapeutic was summarized using NLP techniques. 941 users (29.9%) recorded 31,443 notes. A lexicon of note content was generated from the most common words (Table 1). We tested if note usage was associated with A1C reduction in a subgroup (N = 378) with two lab A1C values. A repeated measures ANCOVA, controlling for demographics and usage, showed the highest note takers exhibiting significantly larger declines in A1C (-2.13%) compared with those who took no notes (-0.86%) or very few notes (-0.79%), F(3,357) = 3.55, p =.02. Conclusions: A substantial subgroup used free-text notes in BlueStar reflecting a diverse range of self-management themes from nondiabetes physical/mental health issues to core diabetes concerns like BG, medication, diet, and activity. Contextual annotation was associated with greater A1C improvement, suggesting that note-taking may offer unique benefit to digital therapeutic users; future studies should address causality. Disclosure M. Dugas: None. K. Crowley: None. W. Wang: None. A.K. Iyer: Employee; Self; WellDoc, Inc. M.M. Peeples: Employee; Self; WellDoc, Inc. M. Shomali: Consultant; Self; WellDoc, Inc. G. Gao: None.

Archive, Edition, Project: Mapping Melville and Networks of Correspondence

Mapping is a critical process of plotting data, images, and textual objects, in space over time, as a way of representing textuality as a kind of continuously evolving landscape of human interaction. A tool for mapping cultural phenomena, like literary coteries and networks, should include ways for users to interact with what is being mapped. It should allow us to make arguments. Digital solutions to these challenges require placing the problem in the context of collaborative and open source technologies and creating a web design that encompasses three areas of development: Archive, Edition, and Projects. In drawing from the experience of working on the Melville Electronic Library (MEL), this essay discusses the potentials and challenges of “mapping Herman Melville.” Both social and geographic mapping can advance scholarship and interpretation, but if they are to work at all and together, they require a digital infrastructure that enables interoperability. The kind of mapping planned for the Melville Electronic Library (MEL) is sustained by a three-part platform: Archive, Edition, and Project. By exploring limits and anxieties related to critical archiving, this essay discusses the inevitability of digital scholarly editing, and strategies for scholarly interaction with archived and edited materials in projects that generate new data (such as biographical and contextual annotation) and return that scholarship to the archive database. Along the way, the essay examines current technologies such as MEL’s editing and transcription tool TextLab, Hofstra DRC’s mapping and annotation tool Itinerary, IIIF image formatting, and Blacklight / Spotlight metadata and content managers.

A Knowledge Fusion Approach for Context Awareness in Vehicular Networks

Vehicular ad-hoc networks (VANETs) are a challenging Internet of Things scenario. While research is proposing increasingly sophisticated hardware and software solutions for on-board context detection, probably high-level context information sharing has not been adequately addressed so far. This paper proposes a novel logic-based framework enabling a contextual data management and mining in VANETs. It grounds on a knowledge fusion algorithm based on nonstandard, nonmonotonic inference services in Description Logics, adopting standard Semantic Web languages. Ontology-referred context annotations produced by individual VANET nodes are merged with automatic reconciliation of inconsistencies. An efficient information dissemination protocol complements the proposal. The approach has been implemented in a vehicular network simulator and early experimental results proved its effectiveness and feasibility.

Exploring Digital Conversation Corpora with Process Mining

Understanding how sentences constitute conversations is still a matter of disagreement. The automatic analysis of corpora can bring new perspectives on this issue. We show that process mining, a computer science approach for analysing human behaviour, enables the inductive identification of within-turn regularities in digital conversations. We use speech acts as a way to discretise conversations and enable the analysis. To this end, we develop a fine-grained, corpus-independent classification of speech acts that we describe and validate. We perform a manual sentence-by-sentence multi-label contextual annotation of a corpus of digital conversations and we explore it with process mining. Our results suggest that speech act theory could be extended to include typicality conditions for sequences of utterances.

Weakly-supervised image captioning based on rich contextual information

Automatically generation of an image description is a challenging task which attracts broad attention in artificial intelligence. Inspired by methods of computer vision and natural language processing, different approaches have been proposed to solve the problem. However, captions generated by the existing approaches have been lack of enough contextual information to describe the corresponding images completely. The labeled captions in the training set only basically describe images and lack of enough contextual annotations. In this paper, we propose a Weakly-supervised Image Captioning Approach (WICA) to generate captions containing rich contextual information, without complete annotations for the contextual information in datasets. We utilize encoder-decoder neural networks to extract basic captioning features and leverage object detection networks to identify contextual features. Then, we encode the two levels of features by a phrase-based language model in order to generate captions with rich contextual information. The comprehensive experimental results reveal that proposed model outperforms the existing baselines in terms of on the richness and reasonability of contextual information for image captioning.

Evidence of selection on splicing-associated loci in human populations and relevance to disease loci mapping

We performed a whole-genome scan of genetic variants in splicing regulatory elements (SREs) and evaluated the extent to which natural selection has shaped extant patterns of variation in SREs. We investigated the degree of differentiation of single nucleotide polymorphisms (SNPs) in SREs among human populations and applied long-range haplotype- and multilocus allelic differentiation-based methods to detect selection signatures. We describe an approach, sampling a large number of loci across the genome from functional classes and using the consensus from multiple tests, for identifying candidates for selection signals. SRE SNPs in various SNP functional classes show different patterns of population differentiation compared with their non-SRE counterparts. Intronic regions display a greater enrichment for extreme population differentiation among the potentially tissue-dependent transcript ratio quantitative trait loci (trQTLs) than SRE SNPs in general and includ outlier trQTLs for cross-population composite likelihood ratio, suggesting that incorporation of context annotation for regulatory variation may lead to improved detection of signature of selection on these loci. The proportion of extremely rare SNPs disrupting SREs is significantly higher in European than in African samples. The approach developed here will be broadly useful for studies of function and disease-associated variation in the human genome.

Nouveaux outils et nouvelles méthodes pour l’enseignement de l’anatomie pathologique. Retour d’expérience à la faculté de médecine Paris Descartes

At the Paris Descartes medicine faculty, we tested some newly developed tools to enhance the pedagogic value of the pathology teaching. In our faculty, this teaching is largely multidisciplinary and integrated in various teaching units; a large part is dedicated to practice works with thirteen 90min sessions. Virtual slides have been used for years in numerous medicine faculties; we successfully implemented this tool by adding contextual annotations, which facilitate students revising. We showed that rewarding students' assiduity enhanced their exam success. To do so, we now propose a short continuous assessment exam at the beginning of each practice session in the form of electronic multi-choice questions. Finally, we now propose a completely computerized final exam, on touchpads, that enhanced its docimologic value.

The BEL information extraction workflow (BELIEF): evaluation in the BioCreative V BEL and IAT track.

Network-based approaches have become extremely important in systems biology to achieve a better understanding of biological mechanisms. For network representation, the Biological Expression Language (BEL) is well designed to collate findings from the scientific literature into biological network models. To facilitate encoding and biocuration of such findings in BEL, a BEL Information Extraction Workflow (BELIEF) was developed. BELIEF provides a web-based curation interface, the BELIEF Dashboard, that incorporates text mining techniques to support the biocurator in the generation of BEL networks. The underlying UIMA-based text mining pipeline (BELIEF Pipeline) uses several named entity recognition processes and relationship extraction methods to detect concepts and BEL relationships in literature. The BELIEF Dashboard allows easy curation of the automatically generated BEL statements and their context annotations. Resulting BEL statements and their context annotations can be syntactically and semantically verified to ensure consistency in the BEL network. In summary, the workflow supports experts in different stages of systems biology network building. Based on the BioCreative V BEL track evaluation, we show that the BELIEF Pipeline automatically extracts relationships with an F-score of 36.4% and fully correct statements can be obtained with an F-score of 30.8%. Participation in the BioCreative V Interactive task (IAT) track with BELIEF revealed a systems usability scale (SUS) of 67. Considering the complexity of the task for new users—learning BEL, working with a completely new interface, and performing complex curation—a score so close to the overall SUS average highlights the usability of BELIEF.Database URL: BELIEF is available at http://www.scaiview.com/belief/

Quantitative Study of Music Listening Behavior in a Smartphone Context

Context-based services have attracted increasing attention because of the prevalence of sensor-rich mobile devices such as smartphones. The idea is to recommend information that a user would be interested in according to the user’s surrounding context. Although remarkable progress has been made to contextualize music playback, relatively little research has been made using a large collection of real-life listening records collected in situ . In light of this fact, we present in this article a quantitative study of the personal, situational, and musical factors of musical preference in a smartphone context, using a new dataset comprising the listening records and self-report context annotation of 48 participants collected over 3wk via an Android app. Although the number of participants is limited and the population is biased towards students, the dataset is unique in that it is collected in a daily context, with sensor data and music listening profiles recorded at the same time. We investigate 3 core research questions evaluating the strength of a rich set of low-level and high-level audio features for music usage auto-tagging (i.e., music preference in different user activities), the strength of time-domain and frequency-domain sensor features for user activity classification, and how user factors such as personality traits are correlated with the predictability of music usage and user activity, using a closed set of 8 activity classes. We provide an in-depth discussion of the main findings of this study and their implications for the development of context-based music services for smartphones.

A system for collecting activity annotations for home energy management

Home energy management is becoming increasingly important and, though there are a plethora of tools for accessing energy consumption data, few provide concrete insights that can directly help users manage demand. Mechanisms that enable a user to draw connections between activities and energy consumption by attaching contextual labels to energy events are a promising step; however solutions for collecting annotations from users can be error prone or intrusive. This work presents a system for collecting in situ annotations using a smartphone application coupled with an off-the-shelf home energy measurement infrastructure. We use a novel power profiling approach to identify important energy consumption events and solicit contextual annotations from the user via a push notification sent to a smartphone. Using a five-week study performed in five homes, we show that our power profiling approach can identify a significant percentage of important energy consumption events using a very small number of monitored devices. We were able to collect an average of over 2 annotations per day and while users provided a wide range of annotations, the motivation to provide annotations varied across subjects.

Discovering contextual tags from product review using semantic relatedness

In the design community, while a number of studies that have focused on studying product reviews in various design analysis perspectives, contextual annotation of identified terms (e.g. product features) has not been fully explored. This paper proposed a learnable approach towards discovering contextual tags from product reviews. A ranking algorithm, FacetRank, is proposed to rank important key terms along with an approach to discover contextual annotation of the terms from review documents. The evaluation of our proposal is performed using two annotated corpus to examine our algorithm’s contextual tagging performance. A case study using a small collection of laptop reviews is also reported to showcase how our algorithm can be applied towards product feature understanding and multi-faceted product ontology development. Finally, we conclude this paper with some indications for future work.

A contextual semantic representation of learning assets in online communities of practice

This paper presents an ontology-based framework for a contextual semantic representation of learning assets within a Community of Practice of E-learning CoPE. The community, made up of actors from the e-learning domain teachers, tutors, pedagogues, administrators , is considered as a virtual space for exchanging and sharing techno-pedagogic knowledge and know-how between those actors. Our objective is to semantically describe the CoPE's learning assets using contextual semantic annotations. We consider two types of semantic annotations: a objective annotations, describing the learning assets with a set of context-related metadata and b subjective annotations, to express the members' experience and feedback regarding these same assets. The paper is illustrated with a case study related to a semantic adaptive wiki using the framework and aiming to foster the knowledge sharing and reuse between the CoPE's members. The wiki provides essentially a semantic search and a recommendation support of assets.

Quantifying and Transferring Contextual Information in Object Detection

Context is critical for reducing the uncertainty in object detection. However, context modeling is challenging because there are often many different types of contextual information coexisting with different degrees of relevance to the detection of target object(s) in different images. It is therefore crucial to devise a context model to automatically quantify and select the most effective contextual information for assisting in detecting the target object. Nevertheless, the diversity of contextual information means that learning a robust context model requires a larger training set than learning the target object appearance model, which may not be available in practice. In this work, a novel context modeling framework is proposed without the need for any prior scene segmentation or context annotation. We formulate a polar geometric context descriptor for representing multiple types of contextual information. In order to quantify context, we propose a new maximum margin context (MMC) model to evaluate and measure the usefulness of contextual information directly and explicitly through a discriminant context inference method. Furthermore, to address the problem of context learning with limited data, we exploit the idea of transfer learning based on the observation that although two categories of objects can have very different visual appearance, there can be similarity in their context and/or the way contextual information helps to distinguish target objects from nontarget objects. To that end, two novel context transfer learning models are proposed which utilize training samples from source object classes to improve the learning of the context model for a target object class based on a joint maximum margin learning framework. Experiments are carried out on PASCAL VOC2005 and VOC2007 data sets, a luggage detection data set extracted from the i-LIDS data set, and a vehicle detection data set extracted from outdoor surveillance footage. Our results validate the effectiveness of the proposed models for quantifying and transferring contextual information, and demonstrate that they outperform related alternative context models.

A unified context assessing model for object categorization

In this paper, we investigate how to appropriately use contextual information for object categorization tasks, especially in the statistical bag-of-features (BoF) framework. Our main contributions are two-folds. Firstly, we propose a context measuring mechanism which could explicitly assess roles of context features for different object recognition tasks. By analyzing information entropy and data ambiguity, only the useful and confident context information would have a final impact on categorization. Secondly, based on the context assessing results, under the BoF framework, we design unified object representations that incorporate the object appearance and contextual information from multiple spatial levels without the need of prior scene segmentations or context annotations. We evaluate the proposed method by the PASCAL object categorization task. The experimental results demonstrate that the proposed context modeling approach improves object categorization significantly and outperforms several state-of-the-art context models.

Poetry and politics in contemporary Bedouin society

This book shows how colloquial Bedouin poetry remains a vibrant art that has manifold modern functions: commenting on world affairs (such as the Arab-Israeli wars, the Gulf War, the American invasion of Iraq); criticising the domestic policies of Arab states; and, highlighting poverty, discrimination, the corrupt practices of officialdom, and compliant local media. Each of the forty-one poems presented within this book is transliterated and translated into English verse, with historical and contextual annotation. The tone is sometimes bitter, sometimes satirical, sometimes scurrilous, and often amusing. The poems are prefaced by an essay on the practice of modern Bedouin poetry. This book is completed by appendices containing the Arabic script versions of the poems, extensive language notes, and a glossary of the vocabulary.

"Contextual annotation framework using ontology based contextual annotation service" by Thanyalak Maneewatthana, with Jessica Rubart as author

Thanyalak Maneewatthana received her PhD from the University of Southampton, UK. The title of her thesis is "Contextual Annotation Framework using Ontology Based Contextual Annotation Service". She currently works at National Statistical Office, Thailand. Her research interests include semantic web technologies, knowledge management, service-oriented architecture.

Taming Data

A challenge in systems-level investigations of the immune response is the principled integration of disparate data sets for constructing predictive models. InnateDB (Lynn et al., 2008; http://www.innatedb.ca), a publicly available, manually curated database of experimentally verified molecular interactions and pathways involved in innate immunity, is a powerful new resource that facilitates such integrative systems-level analyses.

PhotoMap: from location and time to context-aware photo annotations

Despite the growth of geotagged multimedia data on the Web, spatial and temporal metadata are still poorly exploited by Web search engines and multimedia systems. Interpretation and inference processes can enlarge these metadata towards high quality information that are useful for annotating multimedia automatically. In this article, we propose a semi-automatic approach for annotating personal photos based on the use of OWL-DL ontologies together with the next generation of mobile devices. We present an ontology called ContextPhoto and a contextual photo annotation approach which improve the development of more efficient personal image management tools. ContextPhoto provides concepts for representing captured and inferred context information reusing Web standards that describe spatial, temporal and social networking data. In order to validate the context annotation process we propose, we have also designed and developed a mobile and Web location-based system. This new system, called PhotoMap, is an evolution of the related mobile annotation systems since it provides automatic annotation about the spatial, temporal and social contexts of a photo. We also present a demonstration of the PhotoMap application during a tourist tour in the city of Rome.

Influence of contextual information in emotion annotation for spoken dialogue systems

In this paper, we study the impact of considering context information for the annotation of emotions. Concretely, we propose the inclusion of the history of user–system interaction and the neutral speaking style of users. A new method to automatically include both sources of information has been developed making use of novel techniques for acoustic normalization and dialogue context annotation. We have carried out experiments with a corpus extracted from real human interactions with a spoken dialogue system. Results show that the performance of non-expert human annotators and machine-learned classifications are both affected by contextual information. The proposed method allows the annotation of more non-neutral emotions and yields values closer to maximum agreement rates for non-expert human annotation. Moreover, automatic classification accuracy improves by 29.57% compared to the classical approach based only on acoustic features.

Wearable sensing to annotate meeting recordings

We propose to use wearable computers and sensor systems to generate personal contextual annotations in audio visual recordings of meetings. In this paper, we argue that such annotations are essential and effective to allow the retrieval of relevant information from large audio-visual databases. The paper proposes several useful annotations that can be derived from cheap and unobtrusive sensors. It also describes a hardware platform designed to implement accelerometric activity recognition, outlines approaches to extract annotations and presents first experimental results.