Multiple Types Of Entities Research Articles

Every Thing Can Be a Hero! Narrative Visualization of Person, Object, and Other Biographies

Knowledge communication in cultural heritage and digital humanities currently faces two challenges, which this paper addresses: On the one hand, data-driven storytelling in these fields has mainly focused on human protagonists, while other essential entities (such as artworks and artifacts, institutions, or places) have been neglected. On the other hand, storytelling tools rarely support the larger chains of data practices, which are required to generate and shape the data and visualizations needed for such stories. This paper introduces the InTaVia platform, which has been developed to bridge these gaps. It supports the practices of data retrieval, creation, curation, analysis, and communication with coherent visualization support for multiple types of entities. We illustrate the added value of this open platform for storytelling with four case studies, focusing on (a) the life of Albrecht Dürer (person biography), (b) the Saliera salt cellar by Benvenuto Cellini (object biography), (c) the artist community of Lake Tuusula (group biography), and (d) the history of the Hofburg building complex in Vienna (place biography). Numerous suggestions for future research arise from this undertaking.

Urban Economic Structures as Multidimensional Networks: A Complex Systems Framework for Analyzing Economic Development Trajectories

Urban and regional systems often face the difficulty and necessity of structural transitions. These transitions, which can be imposed by external circumstances or initiated by a city itself, include energy transitions, transitions to a circular economy, transitions following a pandemic or natural disaster, or intentional policies meant to “move” an urban economy toward a desired state. However, what does economic structure mean in these cases? Traditional notions of economic structure are ambiguous and simplistic and typically consist of simple distributions, such as number of workers per industry. Yet to better understand, guide, or respond to system transitions, planners must move beyond these nebulous notions toward a theoretically grounded, quantifiable definition of economic structure. A recent trend emerging from the nexus of complexity science and urban science has been to operationalize urban economic structures as networks of interacting economic components. Typically based on colocation patterns of some type of entity, these networks have previously been constructed using economic entities such as products, occupations, or labor skills. Yet different types of entities also exhibit colocation patterns with each other, such as patent technology classes and industries. Here, those cross-entity colocation patterns are used to merge multiple types of entities into a single network representation of urban economies, offering a granularity not possible using a single node type. Occupations, industries, college degrees, and patent technology codes are merged into one multidimensional or multinodal network. As in previous studies, a dense core of highly connected entities emerges in this network. The network locations of individual cities are contrasted, and community detection algorithms are used to identify clusters of highly connected economic entities, showing that the densely connected network core is associated with science, technology, and business-related economic entities. Proximities between individual cities within the network are also measured revealing that many cities that are close to each other in the network are also close to each other in physical space. This framework offers potential applications including the ability to quantify structural change over time in response to a shock or to assess the relative difficulty of future desirable trajectories. More broadly, this framework might be applied to the study of structural change in other complex adaptive systems from human institutions to ecosystems.

Demonstration-Based and Attention-Enhanced Grid-Tagging Network for Mention Recognition

Concepts empower cognitive intelligence. Extracting flat, nested, and discontinuous name entities and concept mentions from natural language texts is significant for downstream tasks such as concept knowledge graphs. Among the algorithms that uniformly detect these types of name entities and concepts, Li et al. proposed a novel architecture by modeling the unified mention recognition as the classification of word–word relations, named W2NER, achieved state-of-the-art (SOTA) results in 2022. However, there is still room for improvement. This paper presents three improvements based on W2NER. We enhanced the grid-tagging network by demonstration learning and tag attention feature extraction, so our modified model is named DTaE. Firstly, addressing the issue of insufficient semantic information in short texts and the lack of annotated data, and inspired by the demonstration learning from GPT-3, a demonstration is searched during the training phase according to a certain strategy to enhance the input features and improve the model’s ability for few-shot learning. Secondly, to tackle the problem of W2NER’s subpar recognition accuracy problem for discontinuous entities and concepts, a multi-head attention mechanism is employed to capture attention scores for different positions based on grid tagging. Then, the tagging attention features are embedded into the model. Finally, to retain information about the sequence position, rotary position embedding is introduced to ensure robustness. We selected an authoritative Chinese dictionary and adopted a five-person annotation method to annotate multiple types of entities and concepts in the definitions. To validate the effectiveness of our enhanced model, experiments were conducted on the public dataset CADEC and our annotated Chinese dictionary dataset: on the CADEC dataset, with a slight decrease in recall rate, precision is improved by 2.78%, and the comprehensive metric F1 is increased by 0.89%; on the Chinese dictionary dataset, the precision is improved by 2.97%, the recall rate is increased by 2.35%, and the comprehensive metric F1 is improved by 2.66%.

Multisource hierarchical neural network for knowledge graph embedding

Link prediction for knowledge graphs aims to obtain missing nodes in triples. In recent years, link prediction methods have made specific achievements in knowledge graph embedding. However, knowledge graphs are characteristic of the heterogeneity of multiple types of entities and relations. A vital issue is efficiently extracting complex graph information and constructing a knowledge-semantic fusion of multiple features. To overcome these issues, a novel link prediction framework based on a multisource hierarchical neural network for knowledge graph embedding (MSHE) is proposed. In particular, mapping functions obtain entities and relations from low- to high-dimensional mapping sources. The combination of mapping sources and entity-relation sources constitutes multisource knowledge information, which facilitates the integration of complex heterogeneous entities and relations. Unlike training a single independent network, the hierarchical embedding network proposed in this paper accumulates feature information at multiple levels. Then, to fuse feature information, our Highway multilayer perceptron (MLP) inductively generates high-quality knowledge information. Through extensive experiments, MSHE’s knowledge graph embedding outperformed the state-of-the-art baselines on FB15k-237 and YAGO3-10. Furthermore, MSHE achieves a Hits@10 score that is 3.8% and 2.7% higher than that of ComplexGCN on FB15K-237 and WN18RR, respectively. MSHE achieves a higher score in Hits@1 than DCN 10.0% in the dataset YAGO3-10. The experiments show that the MSHE achieved excellent results in the four datasets of comparative experiments.

Medical Named Entity Recognition Fusing Part-of-Speech and Stroke Features

It is highly significant from a research standpoint and a valuable practice to identify diseases, symptoms, drugs, examinations, and other medical entities in medical text data to support knowledge maps, question and answer systems, and other downstream tasks that can provide the public with knowledgeable answers. However, when contrasted with other languages like English, Chinese words lack a distinct dividing line, and medical entities have problems such as long length and multiple entity types nesting. Therefore, to address these issues, this study suggests a medical named entity recognition (NER) approach that combines part-of-speech and stroke features. First, the text is fed into the BERT pre-training model to get the semantic representation of the text, while the part-of-speech feature vector is obtained using the part-of-speech dictionary, and the stroke feature of the text is extracted through a convolution neural network (CNN). The word vector is then joined with the part-of-speech and stroke feature vectors, respectively, and input into the BiLSTM and CRF layer for training. Additionally, to balance the disparity in data volume across several types of entities, the class-weighted loss function is included in the loss function. According to the experimental findings, our model’s F1 score on the CCKS2019 dataset reaches 78.65%, and the recognition performance exceeds many existing algorithms.

Revisiting Electrocatalyst Design by a Knowledge Graph of Cu-Based Catalysts for CO2 Reduction

Electrocatalysis takes a significant role in the production of sustainable fuels and chemicals. The combination of artificial intelligence and catalytic science is exhibiting great potential to extract, analyze, and predict electrocatalysts. However, the currently developed machine learning approach usually requires a mass of data from density functional theory calculations to train and optimize models. In contrast, a knowledge graph has the potential to extract useful information from a large amount of the literature without referring to density functional theory. Herein, a knowledge graph of Cu-based electrocatalysts for electrocatalytic CO2 reduction is constructed based on a linguistically enriched SciBERT-based framework. This framework retrieves multiple types of entities including material, regulation method, product, Faradaic efficiency, etc. from 757 scientific literature, generates representations with abundant domain-specific semantic information, and exhibits the capability to deal with electrocatalysts for CO2 reduction. The obtained graph shows the development history of related catalysts, builds relationships between the factors associated with catalysis, and provides intuitive charts for researchers to gain useful information. Furthermore, we propose a deep learning-based prediction model, which integrates the semantic information from the scientific literature (word embedding) with the correlation of knowledge triples (graph embedding) and realizes the prediction of the Faradaic efficiency for a targeted case. This work paves the way for catalyst design in the manner of merging artificial intelligence with catalytic science.

TaughtNet: Learning Multi-Task Biomedical Named Entity Recognition From Single-Task Teachers.

In Biomedical Named Entity Recognition (BioNER), the use of current cutting-edge deep learning-based methods, such as deep bidirectional transformers (e.g. BERT, GPT-3), can be substantially hampered by the absence of publicly accessible annotated datasets. When the BioNER system is required to annotate multiple entity types, various challenges arise because the majority of current publicly available datasets contain annotations for just one entity type: for example, mentions of disease entities may not be annotated in a dataset specialized in the recognition of drugs, resulting in a poor ground truth when using the two datasets to train a single multi-task model. In this work, we propose TaughtNet, a knowledge distillation-based framework allowing us to fine-tune a single multi-task student model by leveraging both the ground truth and the knowledge of single-task teachers. Our experiments on the recognition of mentions of diseases, chemical compounds and genes show the appropriateness and relevance of our approach w.r.t. strong state-of-the-art baselines in terms of precision, recall and F1 scores. Moreover, TaughtNet allows us to train smaller and lighter student models, which may be easier to be used in real-world scenarios, where they have to be deployed on limited-memory hardware devices and guarantee fast inferences, and shows a high potential to provide explainability. We publicly release both our code on github1 and our multi-task model on the huggingface repository.2.

From base data to knowledge discovery – A life cycle approach – Using multilayer networks

Analysis of complex data sets to infer/discover meaningful information/knowledge involves (after data collection and cleaning): (i) Modeling the data — an approach for deriving a suitable representation of data for analysis, (ii) translating analysis objectives into computations on the generated model instance; these computations can be as simple as a query or a complex computation (e.g., community detection over multiple layers), (iii) computation of expressions generated — considering efficiency and scalability, and (iv) drill-down of results to understand them clearly. Beyond this, it is also useful to visualize results for easier understanding. Covid-19 visualization dashboard presented in this paper is an example of this.This paper covers the above steps of data analysis life cycle using a representation (or model) that is gaining importance. With complex data sets containing multiple entity types and relationships, an appropriate model to represent the data is important. For these data sets, we first establish the advantages of Multilayer Networks (or MLNs) as a data model. Then we use an entity-relationship based approach to convert the data set into MLNs for a precise representation of the data set. After that, we outline how expected analysis objectives can be translated using keyword-mapping to aggregate analysis expressions. Finally, we demonstrate, through a set of example data sets and objectives, how the expressions corresponding to objectives are evaluated using an efficient decoupling-based approach. Results are further drilled down to obtain actionable knowledge from the data set.Using the widely popular Enhanced Entity Relationship (EER) approach for requirements representation, we demonstrate how to generate EER diagrams for data sets and further generate, algorithmically, MLNs as well as Relational schema for analysis and drill down, respectively. Using communities and centrality for aggregate analysis, we demonstrate the flexibility of the chosen model to support diverse set of objectives. We also show that compared to current analysis approaches, a “decoupling-based” approach using MLNs is more appropriate as it preserves structure as well as semantics of the results and is very efficient. For this computation, we need to derive expressions for each analysis objective using the MLN model. We provide guidelines to translate English queries into analysis expressions based on keywords.Finally, we use several data sets to establish the effectiveness of modeling using MLNs and their analysis using the decoupling approach that has been proposed recently. For coverage, we use different types of MLNs for modeling, and community and centrality computations for analysis. The data sets used are from US commercial airlines, IMDb (a large international movie data set), the familiar DBLP (or bibliography database), and the Covid-19 data set. Our experimental analyses using the identified steps validate modeling, breadth of objectives that can be computed, and overall versatility of the life cycle approach. Correctness of results is verified, where possible, using independently available ground truth. Furthermore, we demonstrate drill-down that is afforded by this approach (due to structure and semantics preservation) for a better understanding and visualization of results.

BioRED: a rich biomedical relation extraction dataset.

Automated relation extraction (RE) from biomedical literature is critical for many downstream text mining applications in both research and real-world settings. However, most existing benchmarking datasets for biomedical RE only focus on relations of a single type (e.g. protein–protein interactions) at the sentence level, greatly limiting the development of RE systems in biomedicine. In this work, we first review commonly used named entity recognition (NER) and RE datasets. Then, we present a first-of-its-kind biomedical relation extraction dataset (BioRED) with multiple entity types (e.g. gene/protein, disease, chemical) and relation pairs (e.g. gene–disease; chemical–chemical) at the document level, on a set of 600 PubMed abstracts. Furthermore, we label each relation as describing either a novel finding or previously known background knowledge, enabling automated algorithms to differentiate between novel and background information. We assess the utility of BioRED by benchmarking several existing state-of-the-art methods, including Bidirectional Encoder Representations from Transformers (BERT)-based models, on the NER and RE tasks. Our results show that while existing approaches can reach high performance on the NER task (F-score of 89.3%), there is much room for improvement for the RE task, especially when extracting novel relations (F-score of 47.7%). Our experiments also demonstrate that such a rich dataset can successfully facilitate the development of more accurate, efficient and robust RE systems for biomedicine.Availability: The BioRED dataset and annotation guidelines are freely available at https://ftp.ncbi.nlm.nih.gov/pub/lu/BioRED/.

KG-Predict: A knowledge graph computational framework for drug repurposing

The emergence of large-scale phenotypic, genetic, and other multi-model biochemical data has offered unprecedented opportunities for drug discovery including drug repurposing. Various knowledge graph-based methods have been developed to integrate and analyze complex and heterogeneous data sources to find new therapeutic applications for existing drugs. However, existing methods have limitations in modeling and capturing context-sensitive inter-relationships among tens of thousands of biomedical entities. In this paper, we developed KG-Predict: a knowledge graph computational framework for drug repurposing. We first integrated multiple types of entities and relations from various genotypic and phenotypic databases to construct a knowledge graph termed GP-KG. GP-KG was composed of 1,246,726 associations between 61,146 entities. KG-Predict then aggregated the heterogeneous topological and semantic information from GP-KG to learn low-dimensional representations of entities and relations, and further utilized these representations to infer new drug–disease interactions. In cross-validation experiments, KG-Predict achieved high performances [AUROC (the area under receiver operating characteristic) = 0.981, AUPR (the area under precision–recall) = 0.409 and MRR (the mean reciprocal rank) = 0.261], outperforming other state-of-art graph embedding methods. We applied KG-Predict in identifying novel repositioned candidate drugs for Alzheimer’s disease (AD) and showed that KG-Predict prioritized both FDA-approved and active clinical trial anti-AD drugs among the top (AUROC = 0.868 and AUPR = 0.364).

Modeling a construction project in a matrix-based framework for managing requirement changes

Changes are inevitable in a construction project and can impact project’s time, cost, other aspects such as labor productivity, safety, and working relationships. A key challenge of change management is to identify changes and predict their impacts before implementation so that impacts can be minimized. However, different set of entity types in a construction project such as teams, components, deliverables, activities and parameters, as well as complex interdependencies that exist among these entity types, makes difficult to manage changes. To predict changes and their impacts holistically, entity types should be modeled in an integrated manner explicitly. Therefore, objective of the study is to develop a matrix-based framework to predict changes by modeling multiple entity types in a construction project. The proposed matrix-based framework integrates two matrix-based modeling concepts: Axiomatic Design (AD) and Multiple Domain Matrix (MDM). AD principles are utilized for early change identification. MDM principles are used for tracking change propagations across multiple entity types. A schedule risk model is proposed for estimating delays due to changes in entity types. The scope of study is to predict delays due to requirement changes. Framework is validated using data from an ongoing hospital construction project. The findings revealed framework can identify and predict all possible changes accurately.

Community Detection in Fully-Connected Multi-layer Networks Through Joint Nonnegative Matrix Factorization

Modern data analysis and processing tasks typically involve large sets of structured data. Graphs provide a powerful tool to describe the structure of such data, where the entities and the relationships between them are modeled as the nodes and edges of the graph. Traditional single layer network models are insufficient for describing the multiple entity types and modes of interaction encountered in real-world applications. Recently, multi-layer network models, which consider the different types of interactions both within and across layers, have emerged to model these systems. One of the important tools in understanding the topology of these high-dimensional networks is community detection. In this paper, a joint nonnegative matrix factorization approach is proposed to detect the community structure in multi-layer networks. The proposed approach models the multi-layer network as the union of a multiplex and bipartite network and formulates community detection as a regularized optimization problem. This optimization problem simultaneously finds the nonnegative low-rank embedding of the intra- and inter-layer adjacency matrices while minimizing the distance between the two to guarantee pair-wise similarity across embeddings. The proposed approach can detect the community structure for both homogeneous and heterogeneous multi-layer networks and is robust to noise and sparsity. The performance of the proposed approach is evaluated for both simulated and real networks and compared to state-of-the-art methods.

Hyperspherical Variational Co-embedding for Attributed Networks

Network-based information has been widely explored and exploited in the information retrieval literature. Attributed networks, consisting of nodes, edges as well as attributes describing properties of nodes, are a basic type of network-based data, and are especially useful for many applications. Examples include user profiling in social networks and item recommendation in user-item purchase networks. Learning useful and expressive representations of entities in attributed networks can provide more effective building blocks to down-stream network-based tasks such as link prediction and attribute inference. Practically, input features of attributed networks are normalized as unit directional vectors. However, most network embedding techniques ignore the spherical nature of inputs and focus on learning representations in a Gaussian or Euclidean space, which, we hypothesize, might lead to less effective representations. To obtain more effective representations of attributed networks, we investigate the problem of mapping an attributed network with unit normalized directional features into a non-Gaussian and non-Euclidean space. Specifically, we propose a hyperspherical variational co-embedding for attributed networks (HCAN), which is based on generalized variational auto-encoders for heterogeneous data with multiple types of entities. HCAN jointly learns latent embeddings for both nodes and attributes in a unified hyperspherical space such that the affinities between nodes and attributes can be captured effectively. We argue that this is a crucial feature in many real-world applications of attributed networks. Previous Gaussian network embedding algorithms break the assumption of uninformative prior, which leads to unstable results and poor performance. In contrast, HCAN embeds nodes and attributes as von Mises-Fisher distributions, and allows one to capture the uncertainty of the inferred representations. Experimental results on eight datasets show that HCAN yields better performance in a number of applications compared with nine state-of-the-art baselines.

Clustering Heterogeneous Information Network by Joint Graph Embedding and Nonnegative Matrix Factorization

Many complex systems derived from nature and society consist of multiple types of entities and heterogeneous interactions, which can be effectively modeled as heterogeneous information network (HIN). Structural analysis of heterogeneous networks is of great significance by leveraging the rich semantic information of objects and links in the heterogeneous networks. And, clustering heterogeneous networks aims to group vertices into classes, which sheds light on revealing the structure–function relations of the underlying systems. The current algorithms independently perform the feature extraction and clustering, which are criticized for not fully characterizing the structure of clusters. In this study, we propose a learning model by joint <underline>G</underline>raph <underline>E</underline>mbedding and <underline>N</underline>onnegative <underline>M</underline>atrix <underline>F</underline>actorization (aka GEjNMF ), where feature extraction and clustering are simultaneously learned by exploiting the graph embedding and latent structure of networks. We formulate the objective function of GEjNMF and transform the heterogeneous network clustering problem into a constrained optimization problem, which is effectively solved by l 0 -norm optimization. The advantage of GEjNMF is that features are selected under the guidance of clustering, which improves the performance and saves the running time of algorithms at the same time. The experimental results on three benchmark heterogeneous networks demonstrate that GEjNMF achieves the best performance with the least running time compared with the best state-of-the-art methods. Furthermore, the proposed algorithm is robust across heterogeneous networks from various fields. The proposed model and method provide an effective alternative for heterogeneous network clustering.

BDM–MDM integrated solution concept for scheduling information-driven projects

PurposeScheduling in information-driven design phase of construction projects is challenging due to multiple entity types (teams, components, deliverables, activities or parameters) and their dependencies/linkages. Established techniques such as dependency structure matrix (DSM), beeline diagramming method (BDM), multiple domain matrix (MDM), etc. have been independently utilized in past to model information dependencies/linkages and associated iteration. However, there has not been a holistic solution yet for scheduling multiple entity types and their relationships. Hence, an integrated solution needs to be developed that schedules information-driven projects accurately.Design/methodology/approachA case study data collection approach is utilized. With data from two projects, i.e. hostel design and highway design, a BDM–MDM integrated solution was developed and applied to the same. Feedback from experts was obtained for refinements.FindingsThe proposed solution is efficient for scheduling multiple entity types and their information dependencies/linkages.Practical implicationsThe proposed integrated solution enables the project participants to schedule information-driven projects systematically. Application to two distinct design cases emphasizes that the concept is generic and can be applied to any information-driven project with multiple entity types.Originality/valueThe BDM–MDM integrated solution concept is investigated for scheduling multiple entity types in any information-driven projects. This study also explored the terminologies such as multiple entity types and information-driven scheduling.

Link weights recovery in heterogeneous information networks

Socio-technical systems usually consist of many intertwined networks, each connecting different types of objects or actors through a variety of means. As these networks are co-dependent, one can take advantage of this entangled structure to study interaction patterns in a particular network from the information provided by other related networks. A method is, hence, proposed and tested to recover the weights of missing or unobserved links in heterogeneous information networks (HIN)—abstract representations of systems composed of multiple types of entities and their relations. Given a pair of nodes in a HIN, this work aims at recovering the exact weight of the incident link to these two nodes, knowing some other links present in the HIN. To do so, probability distributions resulting from path-constrained random walks, i.e., random walks where the walker is forced to follow only a specific sequence of node types and edge types, capable to capture specific semantics and commonly called a meta-path, are combined in a linearly fashion to approximate the desired result. This method is general enough to compute the link weight between any types of nodes. Experiments on Twitter and bibliographic data show the applicability of the method.

HunFlair: an easy-to-use tool for state-of-the-art biomedical named entity recognition.

SummaryNamed entity recognition (NER) is an important step in biomedical information extraction pipelines. Tools for NER should be easy to use, cover multiple entity types, be highly accurate and be robust toward variations in text genre and style. We present HunFlair, a NER tagger fulfilling these requirements. HunFlair is integrated into the widely used NLP framework Flair, recognizes five biomedical entity types, reaches or overcomes state-of-the-art performance on a wide set of evaluation corpora, and is trained in a cross-corpus setting to avoid corpus-specific bias. Technically, it uses a character-level language model pretrained on roughly 24 million biomedical abstracts and three million full texts. It outperforms other off-the-shelf biomedical NER tools with an average gain of 7.26 pp over the next best tool in a cross-corpus setting and achieves on-par results with state-of-the-art research prototypes in in-corpus experiments. HunFlair can be installed with a single command and is applied with only four lines of code. Furthermore, it is accompanied by harmonized versions of 23 biomedical NER corpora.Availability and implementationHunFlair ist freely available through the Flair NLP framework (https://github.com/flairNLP/flair) under an MIT license and is compatible with all major operating systems.Supplementary informationSupplementary data are available at Bioinformatics online.

Answering Why-Questions for Subgraph Queries

Subgraph queries are routinely used to search for entities in richly attributed graphs e.g., social networks and knowledge graphs. With little knowledge of underlying data, users often need to rewrite queries multiple times to reach desirable answers. Why-questions are studied to clarify missing or unexpected query results. This paper makes a first step to answer Why-questions for entity search in attributed graphs. We consider three common types of Why-questions: Why-not, Why, and Why-rank, which suggest query manipulations that are responsible for user-specified missing, unexpected, and undesirably ranked entities, respectively. (1) We approach a general query rewriting paradigm that suggests to identify desired entities that are specified by Why-questions. We introduce measures that characterize good query rewrites by incorporating both query editing cost and answer closeness. (2) While computing optimal query rewrites is intractable, we develop feasible algorithms, from approximation to fast heuristics, and provide query rewrites with (near) optimality guarantees whenever possible, for Why, Why-not and Why-rank questions. We further show that our results remain intact for Why questions that (1) request a single query rewrite to clarify multiple types of entities, and (2) variants such as Why-empty and Why-so-many, by providing the matching algorithms. Using real-world graphs, we experimentally verify that our algorithms are effective and feasible for large graphs. Our case study also verifies their application in e.g., knowledge exploration.

Entity Resolution in Sparse Encounter Network Using Markov Logic Network

Entity Resolution, which identifies different descriptions referring to the same real-world entity, is a fundamental stage in data integration process essential for quality data analysis. Identities recognition is important in encounter network as it defines the entities of encounters. It is usually not a problem if unique identifier information, e.g., mobile phone number, is available. However, in the circumstances where unique identifier is not available or in question, further investigated is required to perform the entity resolution on the encounter dataset. Often the encounter network is a sparse network with very limited information collected from close-range person-to-person contact reporting, as in epidemiology contact tracing or traffic collision reports. In this paper, we provide an automatic method to resolve the ambiguity of entities in sparse encounter network. We develop a Bayesian spatiotemporal inference system to infer the probability of entity's visits on places of interest. Then, we propose a hierarchical Markov logic network to tackle the inference of the entities in the network which analyses the connection strength of network with multiple types of entities. Experimental results on encounter networks of synthetic and commercial traffic encounter datasets demonstrate that the proposed method achieves better accuracy than existing collective classifications.

Med-Flair: medical named entity recognition for diseases and medications based on Flair embedding

Named Entity Recognition (NER) is a vital step in medical information extraction, especially Electronic Health Records (EHRs). Proper extraction of medical entities such as disease and medications can automate the process of EHR coding as well as considerably improve the filtering of EHR resulting in better extraction of medical information. NER systems are generally trained and evaluated on relatively small standard datasets. However, they are applied on real-world applications, they exposed to different collection of texts, varying in topic, entity distribution, and text type (e.g. abstract vs. full text). This mismatch between the internal structure and application can cause drop in performance and consequently, unreliability. In this paper, we propose Med-Flair, an NER tagger covering mainly multiple entity types, medications and diseases. Med-Flair is mainly based on the Flair NLP framework, in addition, it’s integrated by adding Bidirectional A Long Short Term Memory network (BiLSTM) and Conditional Random Fields (CRF) for sequence tagger. To validate the performance of Med-Flair, it is tested on 4 benchmark datasets, two for medications entities and two for diseases entities. Med Flair successfully achieves high performance, as it achieves 92%, 88%, 92% and 95% of F1-score which are mostly highest compared to state of the art deep neural network architectures such as BioBERT, DTranNER, BERT and BioNerFlair.