Large Knowledge Bases Research Articles

Discussion group networks in occupational medicine: A tool for continuing education to promote the integration of workers with disabilities.

Despite their legal rights, individuals with disabilities face numerous obstacles to integration in the workplace which can result in their discharge from the labor force. Currently occupational physicians have few resources to help decide whether to integrate disabled workers in pre-placement, or in cases of return-to-work. A network of 13 discussion groups comprised of the occupational physicians of each regional clinic of a large Health Maintenance Organization (HMO) in Israel was created to deal with disability management dilemmas. A moderator compiles and shares the physicians' opinions and experiences with all network members thus assisting the consulting physician in decision-making. Successful management of three representative cases is described to illustrate real-life implementations of this network. The network enables both the consulting and other physicians to tap a large knowledge base and decision-making experience concerning cases of occupational disability management, contributing to professional development and improved service delivery.

Computational design of new molecular scaffolds for medicinal chemistry, part II: generalization of analog series-based scaffolds.

Aim:Extending and generalizing the computational concept of analog series-based (ASB) scaffolds.Materials & methods:Methodological modifications were introduced to further increase the coverage of analog series (ASs) and compounds by ASB scaffolds. From bioactive compounds, ASs were systematically extracted and second-generation ASB scaffolds isolated.Results:More than 20,000 second-generation ASB scaffolds with single or multiple substitution sites were extracted from active compounds, achieving more than 90% coverage of ASs.Conclusion:Generalization of the ASB scaffold approach has yielded a large knowledge base of scaffold-capturing compound series and target information.

Incorporating knowledge into neural network for text representation

Text representations is a key task for many natural language processing applications such as document classification, ranking, sentimental analysis and so on. The goal of it is to numerically represent the unstructured text documents so that they can be computed mathematically. Most of the existing methods leverage the power of deep learning to produce a representation of text. However, these models do not consider about the problem that text itself is usually semantically ambiguous and reflects limited information. Due to this reason, it is necessary to seek help from external knowledge base to better understand text.In this paper, we propose a novel framework named Text Concept Vector which leverages both the neural network and the knowledge base to produce a high quality representation of text. Formally, a raw text is primarily conceptualized and represented by a set of concepts through a large taxonomy knowledge base. After that, a neural network is used to transform the conceptualized text into a vector form which encodes both the semantic information and the concept information of the original text. We test our framework on both the sentence level task and the document level task. The experimental results illustrate the effectiveness of our work.

Knowledge-based software design for Defense-in-Depth risk monitor system and application for AP1000

ABSTRACTAs part of the new risk monitor system, the software for the plant Defense-in-Depth (DiD) risk monitor system was designed based on the state-transition and finite-state machine, and then the knowledge-based software was developed by object-oriented method utilizing the Unified Modeling Language (UML). Currently, there are mainly two functions in the developed plant DiD risk monitor software that are knowledge-base editor which is used to model the system in a hierarchical manner and the interaction simulator that simulates the interactions between the different actors in the model. In this paper, a model for playing its behavior is called an Actor which is modeled at the top level. The passive safety AP1000 power plant was studied and the small-break loss-of-coolant accident (SBLOCA) design basis accident transient is modeled using the plant DiD risk monitor software. Furthermore, the simulation result is shown for the interactions between the actors which are defined in the plant DiD risk monitor system as PLANT actor, OPERATOR actor, and SUPERVISOR actor. This paper shows that it is feasible to model the nuclear power plant knowledge base using the software modeling technique. The software can make the large knowledge base for the nuclear power plant with small effort.

Radon – Rapid Discovery of Topological Relations

Geospatial data is at the core of the Semantic Web, of which the largest knowledge base contains more than 30 billions facts. Reasoning on these large amounts of geospatial data requires efficient methods for the computation of links between the resources contained in these knowledge bases. In this paper, we present Radon – efficient solution for the discovery of topological relations between geospatial resources according to the DE9-IM standard. Our evaluation shows that we outperform the state of the art significantly and by several orders of magnitude.

Towards Modeling False Memory With Computational Knowledge Bases.

One challenge to creating realistic cognitive models of memory is the inability to account for the vast common-sense knowledge of human participants. Large computational knowledge bases such as WordNet and DBpedia may offer a solution to this problem but may pose other challenges. This paper explores some of these difficulties through a semantic network spreading activation model of the Deese-Roediger-McDermott false memory task. In three experiments, we show that these knowledge bases only capture a subset of human associations, while irrelevant information introduces noise and makes efficient modeling difficult. We conclude that the contents of these knowledge bases must be augmented and, more important, that the algorithms must be refined and optimized, before large knowledge bases can be widely used for cognitive modeling.

In-database batch and query-time inference over probabilistic graphical models using UDA–GIST

To meet customers' pressing demands, enterprise database vendors have been pushing advanced analytical techniques into databases. Most major DBMSes use user-defined aggregates (UDAs), a data-driven operator, to implement analytical techniques in parallel. However, UDAs alone are not sufficient to implement statistical algorithms where most of the work is performed by iterative transitions over a large state that cannot be naively partitioned due to data dependency. Typically, this type of statistical algorithm requires pre-processing to set up the large state in the first place and demands post-processing after the statistical inference. This paper presents general iterative state transition (GIST), a new database operator for parallel iterative state transitions over large states. GIST receives a state constructed by a UDA and then performs rounds of transitions on the state until it converges. A final UDA performs post-processing and result extraction. We argue that the combination of UDA and GIST (UDA---GIST) unifies data-parallel and state-parallel processing in a single system, thus significantly extending the analytical capabilities of DBMSes. We exemplify the framework through two high-profile batch applications: cross-document coreference, image denoising and one query-time inference application: marginal inference queries over probabilistic knowledge graphs. The 3 applications use probabilistic graphical models, which encode complex relationships of different variables and are powerful for a wide range of problems. We show that the in-database framework allows us to tackle a 27 times larger problem than a scalable distributed solution for the first application and achieves 43 times speedup over the state-of-the-art for the second application. For the third application, we implement query-time inference using the UDA---GIST framework and apply over a probabilistic knowledge graph, achieving 10 times speedup over sequential inference. To the best of our knowledge, this is the first in-database query-time inference engine over large probabilistic knowledge base. We show that the UDA---GIST framework for data- and graph-parallel computations can support both batch and query-time inference efficiently in databases.

ScaLeKB: scalable learning and inference over large knowledge bases

Recent years have seen a drastic rise in the construction of web knowledge bases (e.g., Freebase, YAGO, DBPedia). These knowledge bases store structured information about real-world people, places, organizations, etc. However, due to the limitations of human knowledge, web corpora, and information extraction algorithms, the knowledge bases are still far from complete. To infer the missing knowledge, we propose the Ontological Pathfinding (OP) algorithm to mine first-order inference rules from these web knowledge bases. The OP algorithm scales up via a series of optimization techniques, including a new parallel-rule-mining algorithm, a pruning strategy to eliminate unsound and inefficient rules before applying them, and a novel partitioning algorithm to break the learning task into smaller independent sub-tasks. Combining these techniques, we develop a first rule mining system that scales to Freebase, the largest public knowledge base with 112 million entities and 388 million facts. We mine 36,625 inference rules in 34 h; no existing system achieves this scale. Based on the mining algorithm and the optimizations, we develop an efficient inference engine. As a result, we infer 0.9 billion new facts from Freebase in 17.19 h. We use cross validation to evaluate the inferred facts and estimate a degree of expansion by 0.6 over Freebase, with a precision approaching 1.0. Our approach outperforms state-of-the-art mining algorithms and inference engines in terms of both performance and quality.

Dealing with challenges to methodological pluralism: The paradigm problem, psychological resistance and cultural barriers

This paper calls for methodological pluralism in industrial marketing research. We discuss three challenges that proponents of methodological pluralism have to address if their practice is to be seen as credible: the paradigm problem; psychological resistance; and lack of cultural readiness to accept pluralism. We review the works of a variety of authors from other disciplines who have tackled these problems, and identify useful ideas to take forward into a model of learning. This addresses the paradigm problem by making it clear that no pluralist methodology can exist without making its own paradigmatic assumptions. It deals with psychological resistance by talking in terms of learning, starting from wherever the researcher is currently situated (a large knowledge base is not needed to begin practicing methodological pluralism). However, this model does not deal with the question of whether the time is right, culturally, for methodological pluralism. We argue that the time will be right when it is widely appreciated that methodological pluralism adds value to industrial marketing research practice. The next step for our research community must be the accumulation of a body of empirical evidence to demonstrate that this added value does or does not exist.

Industry use of virtual reality in product design and manufacturing: a survey

In 1999, Fred Brooks, virtual reality pioneer and Professor of Computer Science at the University of North Carolina at Chapel Hill, published a seminal paper describing the current state of virtual reality (VR) technologies and applications (Brooks in IEEE Comput Graph Appl 19(6):16, 1999). Through his extensive survey of industry, Brooks concluded that virtual reality had finally arrived and barely works. His report included a variety of industries which leveraged these technologies to support industry-level innovation. Virtual reality was being employed to empower decision making in design, evaluation, and training processes across multiple disciplines. Over the past two decades, both industrial and academic communities have contributed to a large knowledge base on numerous virtual reality topics. Technical advances have enabled designers and engineers to explore and interact with data in increasingly natural ways. Sixteen years have passed since Brooks original survey. Where are we now? The research presented here seeks to describe the current state of the art of virtual reality as it is used as a decision-making tool in product design, particularly in engineering-focused businesses. To this end, a survey of industry was conducted over several months spanning fall 2014 and spring 2015. Data on virtual reality applications across a variety of industries was gathered through a series of on-site visits. In total, on-site visits with 18 companies using virtual reality were conducted as well as remote conference calls with two others. The authors interviewed 62 people across numerous companies from varying disciplines and perspectives. Success stories and existing challenges were highlighted. While virtual reality hardware has made considerable strides, unique attention was given to applications and the associated decisions that they support. Results suggest that virtual reality has arrived: it works! It is mature, stable, and, most importantly, usable. VR is actively being used in a number of industries to support decision making and enable innovation. Insights from this survey can be leveraged to help guide future research directions in virtual reality technology and applications.

ArchimedesOne

Knowledge bases are becoming increasingly important in structuring and representing information from the web. Meanwhile, web-scale information poses significant scalability and quality challenges to knowledge base systems. To address these challenges, we develop a probabilistic knowledge base system, A rchimedes O ne , by scaling up the knowledge expansion and statistical inference algorithms. We design a web interface for users to query and update large knowledge bases. In this paper, we demonstrate the A rchimedes O ne system to showcase its efficient query and inference engines. The demonstration serves two purposes: 1) to provide an interface for users to interact with A rchimedes O ne through load, search, and update queries; and 2) to validate our approaches of knowledge expansion by applying inference rules in batches using relational operations and query-driven inference by focusing computation on the query facts. We compare A rchimedes O ne with state-of-the-art approaches using two knowledge bases: NELL-sports with 4.5 million facts and Reverb-Sherlock with 15 million facts.

Modeling human daily preferences through a context-aware web-mapping system using semantic technologies

In this paper, a novel geosocial networking service called “G-SPLIS” (Geosocial Semantic Personalized Location Information System) is presented. The paper provides a methodology to design, implement and share in a formal way human daily preferences regarding points of interest (POIs) and POI owners’ group targeted offering policies, via user-defined preferences and policy rules. By adding rules at run time users have more flexibility and they do not rely on the pre-determined application’s methods to get personalized information. Furthermore, G-SPLIS provides a large knowledge base for other systems in the web, because rules are easily sharable. To achieve the above, the presented system is compatible with Semantic Web standards such as the schema.org ontology and uses RuleML for rules that define regular users’ preferences and POI owner’s group-targeted offers. Finally, it combines at run-time the above to match user context with related information and visualizes personalized information.

Abstract 3637: Targeted exome and panel analysis from low input and FFPE DNA using hybridization capture for cancer genome studies

Abstract Introduction Hybridization-based target enrichment techniques coupled with Next Generation Sequencing (NGS) provide a useful and cost-efficient means to study disease specific target regions including whole exomes and gene panels. More than 500 whole exome analyses are behind The Cancer Genome Atlas (TCGA), which is the largest knowledge base for cancer studies including research, prevention, treatment, and care. Clinical samples may be limited in input and of compromised quality due to formalin fixation, however most current NGS library preparation methods require 50-100 ng high quality DNA for such studies. Here we present an efficient method which enables high quality target enrichment and variant calling from inputs as low as 1-25 ng. Method NGS libraries for hybridization capture were made from 1 to 100 ng of Coriell Hapmap samples (NA12878, etc), clinical Formalin Fixed Paraffin Embedded (FFPE) and Horizon Discovery (HDx) reference DNA (HDx 701) using the Accel-NGS® 2S Hyb DNA Library Kit. Amplified libraries were then enriched using specific targeted panels (xGen® Pan-Cancer and xGen® AML) or SeqCap™ EZ MedExome using manufacturer's specifications (IDT™ and Roche NimbleGen™). Enriched libraries were captured using streptavidin beads. Captured libraries were then amplified according to the manufacturer's specifications. Targeted panels were sequenced on an Illumina® MiSeq® using V2 chemistry and MedExome on a HiSeq® using V4 chemistry. Sequence analysis was performed with custom pipelines using BWA for alignment and GATK, Samtools, and Freebayes for variant calling. Results Sequencing analyses yielded a minimum average coverage of 30x and more than thirty fold enrichment. Enriched libraries exhibited significant sequence complexity with minimal duplicates and without any base-composition bias. The percent on-target varied with type of input and ranged from 50-80%. Germline variant calling results had > 99% concordance with the NIST GIAB truth list with sensitivity and precision of > 98%, even from inputs as low as 1 ng of DNA. Somatic variant calling down to 1-5% allele frequency was also evaluated at various DNA input quantities and greater depth of sequencing; results for FFPE and DNA standards will be presented. Conclusions Accel-NGS 2S Hyb technology can be used for whole exome and targeted enrichment studies from low quantity and low quality FFPE clinical samples. High complexity libraries with minimal bias yield high quality sequence data which enables discovery and detection of somatic variants in tumor samples to identify molecular drivers associated with different cancer types. The technique facilitates better understanding of complex cancer genomes and guide precision medicine. Citation Format: Sukhinder K. Sandhu, Cassie Schumacher, Laurie Kurihara, Tim Harkins, Vladimir Makarov. Targeted exome and panel analysis from low input and FFPE DNA using hybridization capture for cancer genome studies. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3637.

MRQUTER: MapReduce 프레임워크를 이용한 병렬 정성 시간 추론기

ABSTRACT In order to meet rapid changes of Web information, it is necessary to extend the current Web technologies to represent both the valid time and location of each fact and knowledge, and reason their relationships. Until recently, many researches on qualitative temporal reasoning have been conducted in laboratory-scale, dealing with small knowledge bases. However, in this paper, we propose the design and implementation of a parallel qualitative temporal reasoner, MRQUTER, which can make reasoning over Web-scale large knowledge bases. This parallel temporal reasoner was built on a Hadoop cluster system using the MapReduce parallel programming framework. It decomposes the entire qualitative temporal reasoning process into several MapReduce jobs such as the encoding and decoding job, the inverse and equal reasoning job, the transitive reasoning job, the refining job, and applies some optimization techniques into each component reasoning job implemented with a pair of Map and Reduce functions. Through experiments using large benchmarking temporal knowledge bases, MRQUTER shows high reasoning performance and scalability.Keywords:Qualitative Temporal Reasoning, Disjunctive Relations, Composition Table, MapReduce, Parallel Temporal Reasoner

Handling failing RDF queries: from diagnosis to relaxation

Recent years have witnessed the development of large knowledge bases (KBs). Due to the lack of information about the content and schema semantics of KBs, users are often not able to correctly formulate KB queries that return the intended result. In this paper, we consider the problem of failing RDF queries, i.e., queries that return an empty set of answers. Query relaxation is one cooperative technique proposed to solve this problem. In the context of RDF data, several works proposed query relaxation operators and ranking models for relaxed queries. But none of them tried to find the causes of an RDF query failure given by Minimal Failing Subqueries (MFSs) as well as successful queries that have a maximal number of triple patterns named Ma$$\underline{x}$$x?imal Succeeding Subqueries (XSSs). Inspired by previous work in the context of relational databases and recommender systems, we propose two complementary approaches to fill this gap. The lattice-based approach (LBA) leverages the theoretical properties of MFSs and XSSs to efficiently explore the subquery lattice of the failing query. The matrix-based approach computes a matrix that records alternative answers to the failing query with the triple patterns they satisfy. The skyline of this matrix directly gives the XSSs of the failing query. This matrix can also be used as an index to improve the performance of LBA. The practical interest of these two approaches are shown via a set of experiments conducted on the LUBM benchmark and a comparative study with baseline and related work algorithms.

An automated knowledge-based textual summarization system for longitudinal, multivariate clinical data

ObjectivesDesign and implement an intelligent free-text summarization system: The system’s input includes large numbers of longitudinal, multivariate, numeric and symbolic clinical raw data, collected over varying periods of time, and in different complex contexts, and a suitable medical knowledge base. The system then automatically generates a textual summary of the data. We aim to prove the feasibility of implementing such a system, and to demonstrate its potential benefits for clinicians and for enhancement of quality of care. MethodsWe have designed a new, domain-independent, knowledge-based system, the CliniText system, for automated summarization in free text of longitudinal medical records of any duration, in any context. The system is composed of six components: (1) A temporal abstraction module generates all possible abstractions from the patient’s raw data using a temporal-abstraction knowledge base; (2) The abductive reasoning module infers abstractions or events from the data, which were not explicitly included in the database; (3) The pruning module filters out raw or abstract data based on predefined heuristics; (4) The document structuring module organizes the remaining raw or abstract data, according to the desired format; (5) The microplanning module, groups the raw or abstract data and creates referring expressions; (6) The surface realization module, generates the text, and applies the grammar rules of the chosen language.We have performed an initial technical evaluation of the system in the cardiac intensive-care and diabetes domains. We also summarize the results of a more detailed evaluation study that we have performed in the intensive-care domain that assessed the completeness, correctness, and overall quality of the system’s generated text, and its potential benefits to clinical decision making. We assessed these measures for 31 letters originally composed by clinicians, and for the same letters when generated by the CliniText system. ResultsWe have successfully implemented all of the components of the CliniText system in software.We have also been able to create a comprehensive temporal-abstraction knowledge base to support its functionality, mostly in the intensive-care domain.The initial technical evaluation of the system in the cardiac intensive-care and diabetes domains has shown great promise, proving the feasibility of constructing and operating such systems.The detailed results of the evaluation in the intensive-care domain are out of scope of the current paper, and we refer the reader to a more detailed source. In all of the letters composed by clinicians, there were at least two important items per letter missed that were included by the CliniText system. The clinicians’ letters got a significantly better grade in three out of four measured quality parameters, as judged by an expert; however, the variance in the quality was much higher in the clinicians’ letters. In addition, three clinicians answered questions based on the discharge letter 40% faster, and answered four out of the five questions equally well or significantly better, when using the CliniText-generated letters, than when using the clinician-composed letters. ConclusionsConstructing a working system for automated summarization in free text of large numbers of varying periods of multivariate longitudinal clinical data is feasible. So is the construction of a large knowledge base, designed to support such a system, in a complex clinical domain, such as the intensive-care domain. The integration of the quality and functionality results suggests that the optimal discharge letter should exploit both human and machine, possibly by creating a machine-generated draft that will be polished by a human clinician.

SlimShot

Increasingly large Knowledge Bases are being created, by crawling the Web or other corpora of documents, and by extracting facts and relations using machine learning techniques. To manage the uncertainty in the data, these KBs rely on probabilistic engines based on Markov Logic Networks (MLN), for which probabilistic inference remains a major challenge. Today's state of the art systems use variants of MCMC, which have no theoretical error guarantees, and, as we show, suffer from poor performance in practice. In this paper we describe SlimShot (Scalable Lifted Inference and Monte Carlo Sampling Hybrid Optimization Technique), a probabilistic inference engine for knowledge bases. SlimShot converts the MLN to a tuple-independent probabilistic database, then uses a simple Monte Carlo-based inference, with three key enhancements: (1) it combines sampling with safe query evaluation, (2) it estimates a conditional probability by jointly computing the numerator and denominator, and (3) it adjusts the proposal distribution based on the sample cardinality. In combination, these three techniques allow us to give formal error guarantees, and we demonstrate empirically that SlimShot outperforms to-day's state of the art probabilistic inference engines used in knowledge bases.

Incremental reasoning on monadic second-order logics with logic programming

Data streams occur widely in various real world applications. The research on streaming data mainly focuses on the data management, query evaluation and optimization on data, but the work on incremental reasoning procedures for streaming knowledge bases is very limited. Typically reasoning services on large knowledge bases are very expensive, and need to be applied continuously when the data is received as a stream. Hence new techniques for optimizing this continuous process is needed for developing efficient reasoners on streaming data. In this paper, we describe a solution to an incremental reasoning problem on an expressive logic, namely monadic second-order logic, and point out further research directions in this area.

Detecting changes to the functioning of a lake ecosystem following a regime shift based on static food-web models

Ecosystem management requires a large base of knowledge regarding the main factors of the components in the system, the relationship between them and their relationships with the external forces that affect ecosystem behavior. In many cases, this knowledge is not always available. Therefore, analyzing previous events and determining the effects they had on the ecosystem can provide insights regarding the factors that caused the changes and their continuing effect on the ecosystem. This task is not simple, due to a large variability of factors in the ecosystem and the difficulty in identifying them. Based on the Ecopath approach, we developed a means for analyzing the Lake Kinneret ecosystem by comparing two mass balance models representing two very different periods. The first model is based on the period of 1990–1993, prior to a regime shift that occurred in the ecosystem and the second model is based on 2006–2010, a period characterized by unstable behavior in the ecosystem. Examining the differences between the two models allowed us to map changes in the ecosystem and to identify the changes and the ecosystem components affected by the regime shift. Using the results we demonstrate the potential of providing management recommendations regarding Lake Kinneret's ecosystem and fishery.

Systematic Assessment of Molecular Selectivity at the Level of Targets, Bioactive Compounds, and Structural Analogues.

Through systematic mining of compound activity data, the target selectivity of bioactive compounds was systematically explored. The analysis was facilitated by applying, extending, and combining the concepts of target cliffs, selectivity cliffs, and matched molecular pairs. Selectivity relationships were explored at different levels including targets, individual bioactive compounds, and pairs of structural analogues. A variety of targets were identified for which active compounds were consistently nonselective or, by contrast, exclusively selective, making it possible to prioritize, or de-prioritize, targets for compound development. Furthermore, many chemical modifications were detected that altered compound selectivity in a well-defined manner including small structural changes that converted nonselective into target-selective compounds or inverted the target selectivity of active compounds. A large knowledge base of selectivity relationships across pharmaceutical targets and chemical modifications that alter selectivity was generated; this has been made freely available to the scientific community as a part of this investigation.