Knowledge Inference Research Articles

Dynamic uncertain causality graph for computer-aided general clinical diagnoses with nasal obstruction as an illustration

Many AI systems have been developed for clinical diagnoses, in which most of them lack interpretability in both knowledge representation and inference results. The newly developed Dynamic Uncertain Causality Graph (DUCG) is a probabilistic graphical model with strong interpretability. However, existing DUCG is mainly for fault diagnoses of large, complex industrial systems. In this paper, we extend DUCG for better application in general clinical diagnoses. Four extensions are introduced: (1) special logic gate and zoom function event variables to represent and quantify the influences of various risk factors on the morbidities of diseases. (2) Reversal logic gate to model the case that some diseases/causes may result in at least two simultaneous symptoms/consequences. (3) Disease-specific manifestation variable for special inference and easy understanding to diagnose a specific disease. (4) Event attention importance to count contributions of isolated state-abnormal variables in inference. To illustrate and verify the extended DUCG methodology, we performed a case study for diagnosing 25 diseases causing nasal obstruction. We tested 171 cases randomly selected from total 471 cases of discharged patients in the hospital information system of Xuanwu Hospital. The diagnosis precision of the extended DUCG was 100%. The diagnosis precision of the third-party verification performed by Suining Central Hospital was 98.86%, which exhibited the strong generalization ability of the extended DUCG.

Perspectives from Ten Years of Protist Studies by High-Throughput Metabarcoding.

During the last decade, high-throughput metabarcoding became routine for analyzing protistan diversity and distributions in nature. Amid a multitude of exciting findings, scientists have also identified and addressed technical and biological limitations, although problems still exist for inference of meaningful taxonomic and ecological knowledge based on short DNA sequences. Given the extensive use of this approach, it is critical to settle our understanding on its strengths and weaknesses and to synthesize up-to-date methodological and conceptual trends. This article summarizes key scientific and technical findings, and identifies current and future directions in protist research that uses metabarcoding.

KNOWLEDGE FROM KNOWLEDGE

Abstract This paper argues that a necessary condition on inferential knowledge is that one knows all the propositions that knowledge depends on. That is, I will argue in support of a principle I call the Knowledge from Knowledge principle: (KFK) S knows that p via inference or reasoning only if S knows all the propositions on which p depends. KFK meshes well with the natural idea that (at least with respect to deductively valid or inductively strong arguments) the epistemic status of one’s belief in the conclusion of an argument is a function of the epistemic status of the attitudes one has to the premises of that argument. One gets what one puts in: S’s belief in the conclusion C of an argument A has epistemic status E for S only if S’s belief in each of the premises C depends on also has epistemic status E. This is compatible, of course, with the epistemic status of S’s conclusion being greater than the epistemic status of the premises on which the conclusion of her reasoning depends. Furthermore, even though KFK applies more straightforwardly to non-suppositional inferences, we can also plausibly apply the principle to sup-positional reasoning if we treat S’s assuming for the sake of argument that p, as S assuming for the sake of argument that she knows that p. Although extremely plausible, philosophers have failed to present a concerted case in support of KFK. Most of their energy goes to articulating and/or defending some form of closure principle for knowledge. What is more, recently versions of this principle have been dismissed as problematic but I think without the reasons in its support been fully appreciated. This paper is an e fort to change the current state of affairs. Its focus is positive, however: my goal is to make the best case I can in support of KFK.⁵ To that end, the paper presents the reasons we have to accept KFK (§1), considers an objection to the principle (§2), and situates the principle in the broader normative context of knowledge norms (§3).

Agent Reliabilism and Inferential Knowledge from Gettiered Belief

AbstractEpistemologists have generally accepted that competently deduced, known conclusions must issue from known premises, as the principle of Counter-Closure demands; however, some have recently challenged the notion, arguing that knowledge may be inferred from non-knowledge. In this paper, I focus on the yet unexamined topic of inferential knowledge from Gettiered belief with regard to Greco's virtue-epistemic framework, which he refers to as ‘agent reliabilism’. I argue that agent reliabilism allows for instances of Counter-Closure violation. In presenting my argument, I construct and provide an analysis of a case that challenges Counter-Closure, and defend the case against various possible objections, including some that feature in mainstream Counter-Closure literature. I contend that, if my analysis is accepted, Greco is left with two options: either he rejects Counter-Closure, or he must substantially revise his view in such a way as to preserve Counter-Closure in light of cases of inferential knowledge from Gettiered belief.

Home service robot task planning using semantic knowledge and probabilistic inference

In the face of unstructured home environment, home service robots are inevitably confronted with uncertainty and incompleteness of environment information. How to make the home service robot obtain enough environment information and plan a discrete sequence of actions through task planning is the key problem of robot intelligence. In this paper, a hierarchical task network based on semantic knowledge and probabilistic inference method is proposed. We use the object location ontology, the location relation between dynamic and static objects to build semantic knowledge of home environment, and build the probability model between dynamic and static objects, as well as between static objects and home scenes. The location of the object is determined by the semantic knowledge and the probability model. Hierarchical task network is selected as an engine of task planner, which can be provided with the location information to improve the autonomy and effectiveness of robot task planning. In order to prevent task execution failure and enhance the adaptability of robot to unstructured home environment, a mechanism of task execution diagnosis and replanning is designed. Experimental results in simulation and real home environment demonstrate that our method can effectively improve the performance of service robot task planning and generate better task execution sequence.

Knowledge graph for identifying hazards on construction sites: Integrating computer vision with ontology

Hazards potentially affect the safety of people on construction sites include falls from heights (FFH), trench and scaffold collapse, electric shock and arc flash/arc blast, and failure to use proper personal protective equipment. Such hazards are significant contributors to accidents and fatalities. Computer vision has been used to automatically detect safety hazards to assist with the mitigation of accidents and fatalities. However, as safety regulations are subject to change and become more stringent prevailing computer vision approaches will become obsolete as they are unable to accommodate the adjustments that are made to practice. This paper integrates computer vision algorithms with ontology models to develop a knowledge graph that can automatically and accurately recognise hazards while adhering to safety regulations, even when they are subjected to change. Our developed knowledge graph consists of: (1) an ontological model for hazards: (2) knowledge extraction; and (3) knowledge inference for hazard identification. We focus on the detection of hazards associated with FFH as an example to illustrate our proposed approach. We also demonstrate that our approach can successfully detect FFH hazards in varying contexts from images.

A new method for interoperability and conformance checking of product manufacturing information

Product manufacturing information (PMI) is the most important information in the product digital model. Whether PMI can be clearly and unambiguous represented and exchanged between different computer-aided software (CAx) has become a topic of high interest in current research. To enhance the interoperability of PMI information among different CAx systems and offer framework software to check the PMI realization ability of different computer-aided design (CAD) software packages, this paper proposes a new method for interoperability and conformance checking by establishing a unified PMI ontology. Using web ontology language description logic (OWL2 DL) and semantic web rule language (SRWL), PMI can be described clearly and without ambiguity and can be read and interpreted automatically by the computer. In addition, the ontology also supplies the functions of consistency checking, knowledge inference, and semantic queries. The analysis shows that this ontology has great engineering significance in realization of PMI interoperability exchange and construction of framework software for PMI conformance checking.

Anlamlı ve Benzer Olmayan Türkçe Metinler Üretmek için N-Gram Yöntemi ile İstatistiksel ve Kural Tabanlı Yaklaşımın Birlikte Kullanımı

Text generation studies are the systems which new knowledge inferences are made by analyzing the existing sentences and meaningful information is obtained from an existing knowledge. These systems provide convenience to users to return more meaningful results related to search results, especially on internet searches. To develop a text generator, there is a need for a linguistic theory to define the sources of natural language and a software tool to process these resources in computer environment. In this study, it is aimed to generate meaningful new Turkish sentences using class-based n-gram model from the sentences in the source data set. Trigram model has been proposed to generate sentences and this model has been developed for use with rule based approach. Unlike other methods, the method used in this study produced meaningful and different sentences with the successive addition method within the framework of the rules determined from the groups divided into triple word groups. Thus, new texts were generated by connecting different sentences from the word or word groups in the source text file as much as the number of the groups that associated with.

A decision support system for demand management in healthcare supply chains considering the epidemic outbreaks: A case study of coronavirus disease 2019 (COVID-19)

The disasters caused by epidemic outbreaks is different from other disasters due to two specific features: their long-term disruption and their increasing propagation. Not controlling such disasters brings about severe disruptions in the supply chains and communities and, thereby, irreparable losses will come into play. Coronavirus disease 2019 (COVID-19) is one of these disasters that has caused severe disruptions across the world and in many supply chains, particularly in the healthcare supply chain. Therefore, this paper, for the first time, develops a practical decision support system based on physicians' knowledge and fuzzy inference system (FIS) in order to help with the demand management in the healthcare supply chain, to reduce stress in the community, to break down the COVID-19 propagation chain, and, generally, to mitigate the epidemic outbreaks for healthcare supply chain disruptions. This approach first divides community residents into four groups based on the risk level of their immune system (namely, very sensitive, sensitive, slightly sensitive, and normal) and by two indicators of age and pre-existing diseases (such as diabetes, heart problems, or high blood pressure). Then, these individuals are classified and are required to observe the regulations of their class. Finally, the efficiency of the proposed approach was measured in the real world using the information from four users and the results showed the effectiveness and accuracy of the proposed approach.

About Territorial Intelligence and Geographic Knowledge Bases

As business intelligence is for companies, territorial intelligence is for local authorities. However, it is not the simple exportation of business intelligence concepts and tools, but overall a new discipline characterized by the amalgamation of collective human intelligence and computer intelligence. As a result, geographic knowledge bases, rules and inference engines are not only based on logic but also on topology, geometry and linguistics, and must take multiple actors’ points of view into account. The scope of this paper will be to present the concepts necessary to construct such systems, i.e. to pass from intelligence to geographic knowledge engineering.

A Survey on Knowledge Graph Embedding: Approaches, Applications and Benchmarks

A knowledge graph (KG), also known as a knowledge base, is a particular kind of network structure in which the node indicates entity and the edge represent relation. However, with the explosion of network volume, the problem of data sparsity that causes large-scale KG systems to calculate and manage difficultly has become more significant. For alleviating the issue, knowledge graph embedding is proposed to embed entities and relations in a KG to a low-, dense and continuous feature space, and endow the yield model with abilities of knowledge inference and fusion. In recent years, many researchers have poured much attention in this approach, and we will systematically introduce the existing state-of-the-art approaches and a variety of applications that benefit from these methods in this paper. In addition, we discuss future prospects for the development of techniques and application trends. Specifically, we first introduce the embedding models that only leverage the information of observed triplets in the KG. We illustrate the overall framework and specific idea and compare the advantages and disadvantages of such approaches. Next, we introduce the advanced models that utilize additional semantic information to improve the performance of the original methods. We divide the additional information into two categories, including textual descriptions and relation paths. The extension approaches in each category are described, following the same classification criteria as those defined for the triplet fact-based models. We then describe two experiments for comparing the performance of listed methods and mention some broader domain tasks such as question answering, recommender systems, and so forth. Finally, we collect several hurdles that need to be overcome and provide a few future research directions for knowledge graph embedding.

A new approach to integrating patient-generated data with expert knowledge for personalized goal setting: A pilot study

IntroductionSelf-monitoring technologies produce patient-generated data that could be leveraged to personalize nutritional goal setting to improve population health; however, most computational approaches are limited when applied to individual-level personalization with sparse and irregular self-monitoring data. We applied informatics methods from expert suggestion systems to a challenging clinical problem: generating personalized nutrition goals from patient-generated diet and blood glucose data. Materials and methodsWe applied qualitative process coding and decision tree modeling to understand how registered dietitians translate patient-generated data into recommendations for dietary self-management of diabetes (i.e., knowledge model). We encoded this process in a set of functions that take diet and blood glucose data as an input and output diet recommendations (i.e., inference engine). Dietitians assessed face validity. Using four patient datasets, we compared our inference engine’s output to clinical narratives and gold standards developed by expert clinicians. ResultsTo dietitians, the knowledge model represented how recommendations from patient data are made. Inference engine recommendations were 63 % consistent with the gold standard (range = 42 %–75 %) and 74 % consistent with narrative clinical observations (range = 63 %–83 %). DiscussionQualitative modeling and automating how dietitians reason over patient data resulted in a knowledge model representing clinical knowledge. However, our knowledge model was less consistent with gold standard than narrative clinical recommendations, raising questions about how best to evaluate approaches that integrate patient-generated data with expert knowledge. ConclusionNew informatics approaches that integrate data-driven methods with expert decision making for personalized goal setting, such as the knowledge base and inference engine presented here, demonstrate the potential to extend the reach of patient-generated data by synthesizing it with clinical knowledge. However, important questions remain about the strengths and weaknesses of computer algorithms developed to discern signal from patient-generated data compared to human experts.

Krishnachandra Bhattacharyya’s Interpolation of Kant’s Idea of the “Self”

Krishnachandra's re-articulation of Kant's transcendental system challenges Kant's conceptualization of 'apperceptive self' conceived as a logical function which is as well the precondition of all our knowledge claims. In Kant's framework, though this "unity of consciousness" is projected as a principle, which undertakes a foundational role as 'apperceptive I', it is capacitated with merely a logical function. Krishnachandra disagrees with Kant's reduction of function of the "self" to a logical process. This reduction would render knowledge of the "self" to be an inferential knowledge, thus making this derivation analogous to the proofs of the transcendental conditions of understanding and sensibility through the logical process of deductions. Krishnachandra's question is: whether this equation established between logical function of 'apperception' and the "self" will suffice to establish the "certitude" of knowledge claims. This is the first task Krishnachandra addresses in his work, Studies in Kant which is elucidated in the following section of this paper. Further, we will see how Krishnachandra’s exploration into the dynamics of this problem leads him to alternatively foreground the "unity", which is much sought by Kantian scholars, between the theoretical and the practical domains of Reason.

PILOT PROJECT FOR ELECTRONIC REIMBURSEMENT SYSTEM FOR PHYSICIANS IN INDONESIA

A healthcare department in remove community of Indonesia aimed for reducing paperwork and improving the electronic system. As part of a pilot project, one aspect was replaced from manual to the electronic format. The proposed system was use of electronic form for claiming for fee reimbursement made by the physicians. The design of the system is intranet based and consisted of two separate portals. The first portal is for physicians and second portal is for billing clerk. The interface is user-friendly and packed with pre-defined codes set in several of its fields and sub-fields. The electronic form is also linked to a centralized database from which a physician can copy the existing patients record. For improving the system variance from individual needs, decision support algorithm is used. Whereas, for improving the system performance, machine learning algorithm is used. For data query, database query was designed. The relationship of columns in the database is displayed as a tabulated form to the user. In situation where a user selects a particular column, a filtered display mechanism displays those columns which satisfying the portion of the query already constructed. For obtaining data from the tabulated database, the SQL query is adapted. Rule-based knowledge inference model is utilized for reasoning about terminology and required domain knowledge. The inference used is algorithmic and helpful in performing all necessary tasks under the suitable billing circumstances. A survey is conducted with 35 physicians for judging their perception towards the system. Results of the survey indicate that most participants find the system suitable and better than the paper-based system in terms of several dimensions such as user friendliness, time saving, reducing errors, and accuracy.

An Integrated Approach of Belief Rule Base and Deep Learning to Predict Air Pollution.

Sensor data are gaining increasing global attention due to the advent of Internet of Things (IoT). Reasoning is applied on such sensor data in order to compute prediction. Generating a health warning that is based on prediction of atmospheric pollution, planning timely evacuation of people from vulnerable areas with respect to prediction of natural disasters, etc., are the use cases of sensor data stream where prediction is vital to protect people and assets. Thus, prediction accuracy is of paramount importance to take preventive steps and avert any untoward situation. Uncertainties of sensor data is a severe factor which hampers prediction accuracy. Belief Rule Based Expert System (BRBES), a knowledge-driven approach, is a widely employed prediction algorithm to deal with such uncertainties based on knowledge base and inference engine. In connection with handling uncertainties, it offers higher accuracy than other such knowledge-driven techniques, e.g., fuzzy logic and Bayesian probability theory. Contrarily, Deep Learning is a data-driven technique, which constitutes a part of Artificial Intelligence (AI). By applying analytics on huge amount of data, Deep Learning learns the hidden representation of data. Thus, Deep Learning can infer prediction by reasoning over available data, such as historical data and sensor data streams. Combined application of BRBES and Deep Learning can compute prediction with improved accuracy by addressing sensor data uncertainties while utilizing its discovered data pattern. Hence, this paper proposes a novel predictive model that is based on the integrated approach of BRBES and Deep Learning. The uniqueness of this model lies in the development of a mathematical model to combine Deep Learning with BRBES and capture the nonlinear dependencies among the relevant variables. We optimized BRBES further by applying parameter and structure optimization on it. Air pollution prediction has been taken as use case of our proposed combined approach. This model has been evaluated against two different datasets. One dataset contains synthetic images with a corresponding label of PM2.5 concentrations. The other one contains real images, PM2.5 concentrations, and numerical weather data of Shanghai, China. We also distinguished a hazy image between polluted air and fog through our proposed model. Our approach has outperformed only BRBES and only Deep Learning in terms of prediction accuracy.

Biomedical Knowledge Graphs Construction From Conditional Statements.

Conditions play an essential role in biomedical statements. However, existing biomedical knowledge graphs (BioKGs) only focus on factual knowledge, organized as a flat relational network of biomedical concepts. These BioKGs ignore the conditions of the facts being valid, which loses essential contexts for knowledge exploration and inference. We consider both facts and their conditions in biomedical statements and proposed a three-layered information-lossless representation of BioKG. The first layer has biomedical concept nodes, attribute nodes. The second layer represents both biomedical fact and condition tuples by nodes of the relation phrases, connecting to the subject and object in the first layer. The third layer has nodes of statements connecting to a set of fact tuples and/or condition tuples in the second layer. We transform the BioKG construction problem into a sequence labeling problem based on a novel designed tag schema. We design a Multi-Input Multi-Output sequence labeling model (MIMO) that learns from multiple input signals and generates proper number of multiple output sequences for tuple extraction. Experiments on a newly constructed dataset show that MIMO outperforms the existing methods. Further case study demonstrates that the BioKGs constructed provide a good understanding of the biomedical statements.

Improving taxonomic relation learning via incorporating relation descriptions into word embeddings

SummaryTaxonomic relations play an important role in various Natural Language Processing (NLP) tasks (eg, information extraction, question answering and knowledge inference). Existing approaches on embedding‐based taxonomic relation learning mainly rely on the word embeddings trained using co‐occurrence‐based similarity learning. However, the performance of these approaches is not quite satisfactory due to the lack of sufficient taxonomic semantic knowledge within word embeddings. To solve this problem, we propose an improved embedding‐based approach to learn taxonomic relations via incorporating relation descriptions into word embeddings. First, to capture additional taxonomic semantic knowledge, we train special word embeddings using not only co‐occurrence information of words but also relation descriptions (eg, taxonomic seed relations and their contextual triples). Then, using the trained word embeddings as features, we employ two learning models to identify and predict taxonomic relations, namely, offset‐based classification model and offset‐based similarity model. Experimental results on four real‐world domain datasets demonstrate that our proposed approach can capture additional taxonomic semantic knowledge and reduce dependence on the training dataset, outperforming the state‐of‐the‐art compared approaches on the taxonomic relation learning task.

Application of knowledge engineering in spacecraft overall design

This paper proposes an advanced technology in spacecraft overall design based on knowledge engineering. Knowledge engineering is a branch of artificial intelligence. Applying knowledge engineering technology to improve the efficiency and quality of overall design process and enhance the automation of overall design, moreover focusing on solving overall spacecraft design knowledge representation, knowledge evaluation, knowledge acquisition, knowledge inference and related issues. Unlike the existing methods, this paper uses ontology technology to build a pint-sized spacecraft overall design knowledge base under protégé software environment, which achieves a better performance. Finally, simulations for the designed power subsystem based on the knowledge base are performed, and the simulation results validate the effectiveness, reliability, of the power subsystem designed by the method of spacecraft overall design based on knowledge engineering.

Dynamic uncertain causality graph based on cloud model theory for knowledge representation and reasoning

The dynamic uncertain causality graph (DUCG), which has been widely applied in many fields, is an important modelling technique for knowledge representation and reasoning. However, the extant DUCG models have been criticized because they cannot precisely represent experts’ knowledge owing to the ignorance of the fuzziness and randomness of uncertain knowledge. In response, we propose a new type of DUCG model called the cloud reasoning dynamic uncertain causality graph (CDUCG). The CDUCG model, which is based on cloud model theory, can handle with the fuzziness and randomness of uncertain information simultaneously. Moreover, an inference algorithm based on the combination of CDUCG and the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) is proposed to implement fuzzy knowledge inference effectively and thus make the expert systems more dependable and intelligent. Finally, illustrative examples and an industrial application concerning root cause analysis of aluminum electrolysis are provided to demonstrate the proposed CDUCG model. And experimental results show that the new CDUCG model is flexible and reliable for knowledge representation and reasoning.

Fusion-Based Cooperative Support Identification for Compressive Networked Sensing

This paper proposes a fusion-based cooperative support identification scheme for distributed compressive sparse signal recovery via resource-constrained wireless sensor networks. The proposed support identification protocol involves: (i) local sparse sensing for economizing data gathering and storage, (ii) local binary decision making for partial support knowledge inference, (iii) binary information exchange among active nodes, and (iv) binary data aggregation for support estimation. Then, with the aid of the estimated signal support, a refined local decision is made at each node. Only the measurements of those informative nodes will be sent to the fusion center, which employs a weighted $\ell_1$-minimization for global signal reconstruction. The design of a Bayesian local decision rule is discussed, and the average communication cost is analyzed. Computer simulations are used to illustrate the effectiveness of the proposed scheme.