Context Of Annotation Research Articles

THÖR-MAGNI: A large-scale indoor motion capture recording of human movement and robot interaction

We present a new large dataset of indoor human and robot navigation and interaction, called THÖR-MAGNI, that is designed to facilitate research on social human navigation: for example, modeling and predicting human motion, analyzing goal-oriented interactions between humans and robots, and investigating visual attention in a social interaction context. THÖR-MAGNI was created to fill a gap in available datasets for human motion analysis and HRI. This gap is characterized by a lack of comprehensive inclusion of exogenous factors and essential target agent cues, which hinders the development of robust models capable of capturing the relationship between contextual cues and human behavior in different scenarios. Unlike existing datasets, THÖR-MAGNI includes a broader set of contextual features and offers multiple scenario variations to facilitate factor isolation. The dataset includes many social human–human and human–robot interaction scenarios, rich context annotations, and multi-modal data, such as walking trajectories, gaze-tracking data, and lidar and camera streams recorded from a mobile robot. We also provide a set of tools for visualization and processing of the recorded data. THÖR-MAGNI is, to the best of our knowledge, unique in the amount and diversity of sensor data collected in a contextualized and socially dynamic environment, capturing natural human–robot interactions.

HyperSOR: Context-Aware Graph Hypernetwork for Salient Object Ranking.

Salient object ranking (SOR) aims to segment salient objects in an image and simultaneously predict their saliency rankings, according to the shifted human attention over different objects. The existing SOR approaches mainly focus on object-based attention, e.g., the semantic and appearance of object. However, we find that the scene context plays a vital role in SOR, in which the saliency ranking of the same object varies a lot at different scenes. In this paper, we thus make the first attempt towards explicitly learning scene context for SOR. Specifically, we establish a large-scale SOR dataset of 24,373 images with rich context annotations, i.e., scene graphs, segmentation, and saliency rankings. Inspired by the data analysis on our dataset, we propose a novel graph hypernetwork, named HyperSOR, for context-aware SOR. In HyperSOR, an initial graph module is developed to segment objects and construct an initial graph by considering both geometry and semantic information. Then, a scene graph generation module with multi-path graph attention mechanism is designed to learn semantic relationships among objects based on the initial graph. Finally, a saliency ranking prediction module dynamically adopts the learned scene context through a novel graph hypernetwork, for inferring the saliency rankings. Experimental results show that our HyperSOR can significantly improve the performance of SOR.

Construction of Algorithm for English Learners to Read Foreign Literature Based on Machine Translation Technology

This study focuses on the development of an algorithm aimed at assisting English learners in comprehending foreign literature through the utilization of machine translation technology. With the growing importance of multilingual proficiency in a globalized world, the ability to read and understand literature in foreign languages is increasingly valued. However, language barriers often hinder learners' comprehension and appreciation of such texts. To address this challenge, our research endeavors to construct an algorithm that leverages machine translation techniques to facilitate the reading process for English learners. By integrating advanced translation algorithms and natural language processing methods, the proposed system aims to provide accurate and contextually relevant translations of foreign literature into English. Additionally, the algorithm will incorporate features tailored to the needs of English learners, such as vocabulary assistance, grammatical explanations, and cultural context annotations, to enhance their reading experience and comprehension levels. Through rigorous experimentation and evaluation, the effectiveness and usability of the algorithm will be assessed, with the ultimate goal of empowering English learners to engage more deeply with foreign literature and broaden their linguistic and cultural horizons.

VarSCAT: A computational tool for sequence context annotations of genomic variants.

The sequence contexts of genomic variants play important roles in understanding biological significances of variants and potential sequencing related variant calling issues. However, methods for assessing the diverse sequence contexts of genomic variants such as tandem repeats and unambiguous annotations have been limited. Herein, we describe the Variant Sequence Context Annotation Tool (VarSCAT) for annotating the sequence contexts of genomic variants, including breakpoint ambiguities, flanking bases of variants, wildtype/mutated DNA sequences, variant nomenclatures, distances between adjacent variants, tandem repeat regions, and custom annotation with user customizable options. Our analyses demonstrate that VarSCAT is more versatile and customizable than the currently available methods or strategies for annotating variants in short tandem repeat (STR) regions or insertions and deletions (indels) with breakpoint ambiguity. Variant sequence context annotations of high-confidence human variant sets with VarSCAT revealed that more than 75% of all human individual germline and clinically relevant indels have breakpoint ambiguities. Moreover, we illustrate that more than 80% of human individual germline small variants in STR regions are indels and that the sizes of these indels correlated with STR motif sizes. VarSCAT is available from https://github.com/elolab/VarSCAT.

MirDIP 5.2: tissue context annotation and novel microRNA curation.

MirDIP is a well-established database that aggregates microRNA-gene human interactions from multiple databases to increase coverage, reduce bias, and improve usability by providing an integrated score proportional to the probability of the interaction occurring. In version 5.2, we removed eight outdated resources, added a new resource (miRNATIP), and ran five prediction algorithms for miRBase and mirGeneDB. In total, mirDIP 5.2 includes 46 364 047 predictions for 27 936 genes and 2734 microRNAs, making it the first database to provide interactions using data from mirGeneDB. Moreover, we curated and integrated 32 497 novel microRNAs from 14 publications to accelerate the use of these novel data. In this release, we also extend the content and functionality of mirDIP by associating contexts with microRNAs, genes, and microRNA-gene interactions. We collected and processed microRNA and gene expression data from 20 resources and acquired information on 330 tissue and disease contexts for 2657 microRNAs, 27 576 genes and 123 651 910 gene-microRNA-tissue interactions. Finally, we improved the usability of mirDIP by enabling the user to search the database using precursor IDs, and we integrated miRAnno, a network-based tool for identifying pathways linked to specific microRNAs. We also provide a mirDIP API to facilitate access to its integrated predictions. Updated mirDIP is available at https://ophid.utoronto.ca/mirDIP.

Passive and Context-Aware In-Home Vital Signs Monitoring Using Co-Located UWB-Depth Sensor Fusion

Basic vital signs such as heart and respiratory rates (HR and RR) are essential bio-indicators. Their longitudinal in-home collection enables prediction and detection of disease onset and change, providing for earlier health intervention. In this article, we propose a robust, non-touch vital signs monitoring system using a pair of co-located Ultra-Wide Band (UWB) and depth sensors. By extensive manual examination, we identify four typical temporal and spectral signal patterns and their suitable vital sign estimators. We devise a probabilistic weighted framework (PWF) that quantifies evidence of these patterns to update the weighted combination of estimator output to track the vital signs robustly. We also design a “heatmap”-based signal quality detector to exclude the disturbed signal from inadvertent motions. To monitor multiple co-habiting subjects in-home, we build a two-branch long short-term memory (LSTM) neural network to distinguish between individuals and their activities, providing activity context crucial to disambiguating critical from normal vital sign variability. To achieve reliable context annotation, we carefully devise the feature set of the consecutive skeletal poses from the depth data, and develop a probabilistic tracking model to tackle non-line-of-sight (NLOS) cases. Our experimental results demonstrate the robustness and superior performance of the individual modules as well as the end-to-end system for passive and context-aware vital sign monitoring.

MetaRNN: differentiating rare pathogenic and rare benign missense SNVs and InDels using deep learning

Multiple computational approaches have been developed to improve our understanding of genetic variants. However, their ability to identify rare pathogenic variants from rare benign ones is still lacking. Using context annotations and deep learning methods, we present pathogenicity prediction models, MetaRNN and MetaRNN-indel, to help identify and prioritize rare nonsynonymous single nucleotide variants (nsSNVs) and non-frameshift insertion/deletions (nfINDELs). We use independent test sets to demonstrate that these new models outperform state-of-the-art competitors and achieve a more interpretable score distribution. Importantly, prediction scores from both models are comparable, enabling easy adoption of integrated genotype-phenotype association analysis methods. All pre-computed nsSNV scores are available at http://www.liulab.science/MetaRNN. The stand-alone program is also available at https://github.com/Chang-Li2019/MetaRNN.

Study of the Automatic Recognition of Landslides by Using InSAR Images and the Improved Mask R-CNN Model in the Eastern Tibet Plateau

The development of landslide hazards is spatially scattered, temporally random, and poorly characterized. Given the advantages of the large spatial scale and high sensitivity of InSAR observations, InSAR is becoming one of the main techniques for active landslide identification. The difficult problem is how to quickly extract landslide information from extensive InSAR image data. Since the instance segmentation model (Mask R-CNN) in deep learning can provide highly robust target recognition, we select the landslide-prone eastern edge of the Tibetan Plateau as a specific test area. Introducing and optimizing this model achieves high-speed and accurate recognition of InSAR observations. First, the InSAR patch landslide instance segmentation dataset (SLD) is established by developing a common object in context (COCO) annotation format conversion code based on InSAR observations. The Mask R-CNN+++ is found by adding three functions of the ResNext module to increase the fineness of the network segmentation results and enhance the noise resistance of the model, the DCB (deformable convolutional block) to improve the feature extraction ability of the network for geometric morphological changes of landslide patches, and an attention mechanism to selectively enhance usefully and suppress features less valuable to the native Mask R-CNN network. The model achieves 92.94% accuracy on the test set, and the active landslide recognition speed based on this model under ordinary computer hardware conditions is 72.3 km2/s. The overall characteristics of the results of this study show that the optimized model effectively enhances the perceptibility of image morphological changes, thereby resulting in smoother recognition boundaries and further improvement of the generalization ability of segmentation detection. This result is expected to serve to identify and monitor active landslides in complex surface conditions on a large spatial scale. Moreover, active landslides of different geometric features, motion patterns, and intensities are expected to be further segmented.

Non-Coding RNAs Are Brokers in Breast Cancer Interactome Networks and Add Discrimination Power between Subtypes.

Despite the power of high-throughput genomics, most non-coding RNA (ncRNA) biotypes remain hard to identify, characterize, and validate. This is a clear indication that intensive next-generation sequencing research has led to great efficiency and accuracy in detecting ncRNAs, but not in their functionalization. Computational scientists continue to support the discovery process by spotting significant data features (expression or mutational profiles), elucidating phenotype uncertainty, and delineating complex regulation landscapes for biological pathways and pathophysiological processes. With reference to transcriptome regulation dynamics in cancer, this work introduces a novel network-driven inference approach designed to reveal the potential role of computationally identified ncRNAs in discriminating between breast cancer (BC) subtypes beyond the traditional gene expression signatures. As heterogeneity cast in the subtypes is a characteristic of most cancers, the proposed approach is generalizable beyond BC. Expression profiles of a wide transcriptome spectrum were obtained for a number of BC patients (and controls) listed in TCGA and processed with RNA-Seq. The well-known PAM50 subtype signature was available for the samples and used to move from differentially expressed transcript profiles to subtype-specific biclusters associating gene patterns with patients. Co-expressed gene networks were then generated and annotations were provided, focusing on the biclusters with basal and luminal signatures. These were used to build template maps, i.e., networks in which to embed the ncRNAs and contextually functionalize them based on their interactors. This inference approach is able to assess the influence of ncRNAs at the level of BC subtype. Network topology was considered through the brokerage measure to account for disruptiveness effects induced by the removal of nodes corresponding to ncRNAs. Equivalently, it is shown that ncRNAs can act as brokers of network interactome dynamics, and removing them allows the refinement of subtype-related characteristics previously obtained by gene signatures only. The results of the study elucidate the role of pseudogenes in two major BC subtypes, considering the contextual annotations. Put into a wider perspective, ncRNA brokers may help predictive functionalization studies targeted to new disease phenotypes, for instance those linked to the tumor microenvironment or metabolism, or those specifically involving metastasis. Overall, the approach may represent an in silico prioritization strategy toward the systems identification of new diagnostic and prognostic biomarkers.

Panoptic Segmentation Meets Remote Sensing

Panoptic segmentation combines instance and semantic predictions, allowing the detection of countable objects and different backgrounds simultaneously. Effectively approaching panoptic segmentation in remotely sensed data is very promising since it provides a complete classification, especially in areas with many elements as the urban setting. However, some difficulties have prevented the growth of this task: (a) it is very laborious to label large images with many classes, (b) there is no software for generating DL samples in the panoptic segmentation format, (c) remote sensing images are often very large requiring methods for selecting and generating samples, and (d) most available software is not friendly to remote sensing data formats (e.g., TIFF). Thus, this study aims to increase the operability of panoptic segmentation in remote sensing by providing: (1) a pipeline for generating panoptic segmentation datasets, (2) software to create deep learning samples in the Common Objects in Context (COCO) annotation format automatically, (3) a novel dataset, (4) leverage the Detectron2 software for compatibility with remote sensing data, and (5) evaluate this task on the urban setting. The proposed pipeline considers three inputs (original image, semantic image, and panoptic image), and our software uses these inputs alongside point shapefiles to automatically generate samples in the COCO annotation format. We generated 3400 samples with 512 × 512 pixel dimensions and evaluated the dataset using Panoptic-FPN. Besides, the metric analysis considered semantic, instance, and panoptic metrics, obtaining 93.865 mean intersection over union (mIoU), 47.691 Average (AP) Precision, and 64.979 Panoptic Quality (PQ). Our study presents the first effective pipeline for generating panoptic segmentation data for remote sensing targets.

In the Hallways of the Tauride Palace during the February Days of 1917: Memoirs of a Veteran Staffer of the Duma Chancellery А. А. Kondrat’ev

This publication presents an autobiographical account of the February Days in Petrograd by a long-standing mid-level staffer of the Duma Chancellery, Aleksandr Alekseevich Kondrat’ev (1876–1967). The author spent the first days of the February Revolution inside the Tauride Palace and diligently recorded his impressions, some of which he also shared in letters to his contemporaries. Later, he systematized what he had remembered and turned it into a cohesive narrative published below. His reflections on the revolution’s key moments offer a unique perspective of a well-informed insider who at the same time remains detached from party politics and allegiances. It is from inside the Tauride Palace, from its offices and hallways that the author witnessed the meteoric rise (and soon, the precipitous downfall) of the propertied but nevertheless revolutionary Fourth Duma, as well as the breakdown of the old political and social order. His testimony augments some of the established views on the attitudes and behavior of contemporaries caught up in the revolutionary whirlwind. Of special interest is the author’s testimony about his fellow Duma Chancellery staffers who — unlike insurgents and Duma politicians who flocked into the revolutionary headquarters in the wake of the unprecedented popular uprising that engulfed the capital city — came to the Tauride Palace on 27 February to fulfill their bureaucratic duty, that is to ensure orderly functioning of the Duma apparatus. The text below is supplemented by textual and contextual annotations incorporating the most up-to-date scholarship.

IID 2021: towards context-specific protein interaction analyses by increased coverage, enhanced annotation and enrichment analysis.

Improved bioassays have significantly increased the rate of identifying new protein-protein interactions (PPIs), and the number of detected human PPIs has greatly exceeded early estimates of human interactome size. These new PPIs provide a more complete view of disease mechanisms but precise understanding of how PPIs affect phenotype remains a challenge. It requires knowledge of PPI context (e.g. tissues, subcellular localizations), and functional roles, especially within pathways and protein complexes. The previous IID release focused on PPI context, providing networks with comprehensive tissue, disease, cellular localization, and druggability annotations. The current update adds developmental stages to the available contexts, and provides a way of assigning context to PPIs that could not be previously annotated due to insufficient data or incompatibility with available context categories (e.g. interactions between membrane and cytoplasmic proteins). This update also annotates PPIs with conservation across species, directionality in pathways, membership in large complexes, interaction stability (i.e. stable or transient), and mutation effects. Enrichment analysis is now available for all annotations, and includes multiple options; for example, context annotations can be analyzed with respect to PPIs or network proteins. In addition to tabular view or download, IID provides online network visualization. This update is available at http://ophid.utoronto.ca/iid.

Putting human behavior predictability in context

Various studies have investigated the predictability of different aspects of human behavior such as mobility patterns, social interactions, and shopping and online behaviors. However, the existing researches have been often limited to a single or to the combination of few behavioral dimensions, and they have adopted the perspective of an outside observer who is unaware of the motivations behind the specific behaviors or activities of a given individual. The key assumption of this work is that human behavior is deliberated based on an individual’s own perception of the situation that s/he is in, and that therefore it should also be studied under the same perspective. Taking inspiration from works in ubiquitous and context-aware computing, we investigate the role played by four contextual dimensions (or modalities), namely time, location, activity being carried out, and social ties, on the predictability of individuals’ behaviors, using a month of collected mobile phone sensor readings and self-reported annotations about these contextual modalities from more than two hundred study participants. Our analysis shows that any target modality (e.g. location) becomes substantially more predictable when information about the other modalities (time, activity, social ties) is made available. Multi-modality turns out to be in some sense fundamental, as some values (e.g. specific activities like “shopping”) are nearly impossible to guess correctly unless the other modalities are known. Subjectivity also has a substantial impact on predictability. A location recognition experiment suggests that subjective location annotations convey more information about activity and social ties than objective information derived from GPS measurements. We conclude the paper by analyzing how the identified contextual modalities allow to compute the diversity of personal behavior, where we show that individuals are more easily identified by rarer, rather than frequent, context annotations. These results offer support in favor of developing innovative computational models of human behaviors enriched by a characterization of the context of a given behavior.

Instance segmentation of center pivot irrigation systems using multi-temporal SENTINEL-1 SAR images

The mapping of Center Pivot Irrigation Systems (CPIS) is essential for agricultural and water resource management. In this context, methods based on Deep Learning (DL) have reached state-of-the-art in the classification of remote sensing images. However, the mapping of CPIS with DL is still restricted to optical images with limitations in tropical environments due to the extensive cloud cover for long periods. The present research proposes the detection of CPIS using instance segmentation from multi-temporal SAR images that are cloud-free. The research developed a CPIS database for the Cerrado biome based on visual interpretation, totaling 3675 instances in the Common Objects in Context (COCO) annotation format. The training used the Mask-RCNN with the ResNeXt-101-32x8d backbone considering different data arrangements: (a) variation in the number of Sentinel-1 temporal images with an interval of 12 days (from 1 to 11 images), and (b) comparison of VV, VH, and VV + VH polarizations. For mapping large areas, we applied mosaicking with a sliding window technique. The results show an accuracy improvement with the increase in the number of temporal images, reaching a difference greater than 15% AP when comparing a single temporal image with the optimal number of temporal images in the VV (eight), VH (ten) and VV + VH (nine) polarizations. The combined use of the two polarizations (VV + VH) had slightly better results (75% AP, 91% AP50, and 86% AP75) than the others. However, VV polarization may have an advantage, obtaining close results from less image and computational cost. The instance segmentation with the sliding window provides the automatic identification of different objects belonging to the same class in large areas, allowing the total CPIS count and the size calculation.

Mol* Viewer: modern web app for 3D visualization and analysis of large biomolecular structures.

Large biomolecular structures are being determined experimentally on a daily basis using established techniques such as crystallography and electron microscopy. In addition, emerging integrative or hybrid methods (I/HM) are producing structural models of huge macromolecular machines and assemblies, sometimes containing 100s of millions of non-hydrogen atoms. The performance requirements for visualization and analysis tools delivering these data are increasing rapidly. Significant progress in developing online, web-native three-dimensional (3D) visualization tools was previously accomplished with the introduction of the LiteMol suite and NGL Viewers. Thereafter, Mol* development was jointly initiated by PDBe and RCSB PDB to combine and build on the strengths of LiteMol (developed by PDBe) and NGL (developed by RCSB PDB). The web-native Mol* Viewer enables 3D visualization and streaming of macromolecular coordinate and experimental data, together with capabilities for displaying structure quality, functional, or biological context annotations. High-performance graphics and data management allows users to simultaneously visualise up to hundreds of (superimposed) protein structures, stream molecular dynamics simulation trajectories, render cell-level models, or display huge I/HM structures. It is the primary 3D structure viewer used by PDBe and RCSB PDB. It can be easily integrated into third-party services. Mol* Viewer is open source and freely available at https://molstar.org/.

Instance Segmentation for Large, Multi-Channel Remote Sensing Imagery Using Mask-RCNN and a Mosaicking Approach

Instance segmentation is the state-of-the-art in object detection, and there are numerous applications in remote sensing data where these algorithms can produce significant results. Nevertheless, one of the main problems is that most algorithms use Red, Green, and Blue (RGB) images, whereas Satellite images often present more channels that can be crucial to improve performance. Therefore, the present work brings three contributions: (a) conversion system from ground truth polygon data into the Creating Common Object in Context (COCO) annotation format; (b) Detectron2 software source code adaptation and application on multi-channel imagery; and (c) large scene image mosaicking. We applied the procedure in a Center Pivot Irrigation System (CPIS) dataset with ground truth produced by the Brazilian National Water Agency (ANA) and Landsat-8 Operational Land Imager (OLI) imagery (7 channels with 30-m resolution). Center pivots are a modern irrigation system technique with massive growth potential in Brazil and other world areas. The round shapes with different textures, colors, and spectral behaviors make it appropriate to use Deep Learning instance segmentation. We trained the model using 512 × 512-pixel sized patches using seven different backbone structures (ResNet50- Feature Pyramid Network (FPN), Resnet50-DC5, ResNet50-C4, Resnet101-FPN, Resnet101-DC5, ResNet101-FPN, and ResNeXt101-FPN). The model evaluation used standard COCO metrics (Average Precision (AP), AP50, AP75, APsmall, APmedium, and AR100). ResNeXt101-FPN had the best results, with a 3% advantage over the second-best model (ResNet101-FPN). We also compared the ResNeXt101-FPN model in the seven-channel and RGB imagery, where the multi-channel model had a 3% advantage, demonstrating great improvement using a larger number of channels. This research is also the first with a mosaicking algorithm using instance segmentation models, where we tested in a 1536 × 1536-pixel image using a non-max suppression sorted by area method. The proposed methodology is innovative and suitable for many other remote sensing problems and medical imagery that often present more channels.

Engagement and Outcomes Associated with Contextual Annotation Features of a Digital Health Solution.

Digital health solutions targeting diabetes self-care are popular and promising, but important questions remain about how these tools can most effectively help patients. Consistent with evidence of the salutary effects of note-taking in education, features that enable annotation of structured data entry might enhance the meaningfulness of the interaction, thereby promoting persistent use and benefits of a digital health solution. To examine the potential benefits of note-taking, we explored how patients with type 2 diabetes used annotation features of a digital health solution and assessed the relationship between annotation and persistence in engagement as well as improvements in glycated hemoglobin (A1C). Secondary data from 3142 users of the BlueStar digital health solution collected between December 2013 and June 2017 were analyzed, with a subgroup of 372 reporting A1C lab values. About a third of patients recorded annotations while using the platform. Annotation themes largely reflected self-management behaviors (diet, physical activity, medication adherence) and well-being (mood, health status). Early use of contextual annotations was associated with greater engagement over time and with greater improvements in A1C. Our research provides preliminary evidence of the benefits of annotation features in a digital health solution. Future research is needed to assess the causal impact of note-taking and the moderating role of thematic content reflected in notes.

Meta-analysis of the coral environmental stress response: Acropora corals show opposing responses depending on stress intensity.

As climate change progresses, reef-building corals must contend more often with suboptimal conditions, motivating a need to understand coral stress response. Here, we test the hypothesis that there is a stereotyped transcriptional response that corals enact under all stressful conditions, functionally characterized by downregulation of growth, and activation of cell death, response to reactive oxygen species, immunity, and protein folding and degradation. We analyse RNA-seq and Tag-Seq data from 14 previously published studies and supplement them with four new experiments involving different stressors, totaling over 600 gene expression profiles from the genus Acropora. Contrary to expectations, we found not one, but two distinct types of response. The type A response was observed under all kinds of high-intensity stress, was correlated between independent projects and was functionally consistent with the hypothesized stereotyped response. The consistent correlation between projects, irrespective of stress type, supports the type A response as the general coral environmental stress response (ESR), a blanket solution to severely stressful conditions. The distinct type B response was observed under lower intensity stress and was more variable among studies. Unexpectedly, at the level of individual genes and functional categories, the type B response was broadly opposite the type A response. Finally, taking advantage of the breadth of the data set, we present contextual annotations for previously unannotated genes based on consistent stress-induced differences across independent projects.

How Emotional and Contextual Annotations Involve in Sensemaking Processes of Foreign Language Social Media Posts

The goal of this paper is to investigate how computational tools to annotate communication can support multilingual sense-making on social media. We conducted a field study of SenseTrans, a browser extension that uses sentiment analysis and named entity extraction techniques to annotate Facebook posts with emotional and contextual information. Interviews with 18 participants who used SenseTrans in their Facebook newsfeed for two weeks suggest that the annotations often supported sensemaking by providing additional information they could use to get a quick gist of the posts or to supplement their own interpretations. Participants varied in the extent to which they were motivated to evaluate the credibility of and form mental models of how the annotations were generated, which shaped how they utilized the annotations for sensemaking. Our findings demonstrate the value of designing to support cross-lingual communication and inform design implications for intelligent tools that support communication and sensemaking.

Assisting Forensic Identification through Unsupervised Information Extraction of Free Text Autopsy Reports: The Disappearances Cases during the Brazilian Military Dictatorship

Anthropological, archaeological, and forensic studies situate enforced disappearance as a strategy associated with the Brazilian military dictatorship (1964–1985), leaving hundreds of persons without identity or cause of death identified. Their forensic reports are the only existing clue for people identification and detection of possible crimes associated with them. The exchange of information among institutions about the identities of disappeared people was not a common practice. Thus, their analysis requires unsupervised techniques, mainly due to the fact that their contextual annotation is extremely time-consuming, difficult to obtain, and with high dependence on the annotator. The use of these techniques allows researchers to assist in the identification and analysis in four areas: Common causes of death, relevant body locations, personal belongings terminology, and correlations between actors such as doctors and police officers involved in the disappearances. This paper analyzes almost 3000 textual reports of missing persons in São Paulo city during the Brazilian dictatorship through unsupervised algorithms of information extraction in Portuguese, identifying named entities and relevant terminology associated with these four criteria. The analysis allowed us to observe terminological patterns relevant for people identification (e.g., presence of rings or similar personal belongings) and automate the study of correlations between actors. The proposed system acts as a first classificatory and indexing middleware of the reports and represents a feasible system that can assist researchers working in pattern search among autopsy reports.