Wyszukiwarka
Wyniki wyszukiwania dla: INFERENCE
-
THE ROLE OF INFERENCE IN MOBILE MEDICAL APPLICATION DESIGN
PublikacjaIn the early 21st century, artificial intelligence began to be used to process medical information. However, before this happened, predictive models used in healthcare could only consider a limited number of variables, and only in properly structured and organised medical data. Today, advanced tools based on machine learning techniques - which, using artificial neural networks, can explore extremely complex relationships - and...
-
Journal of Causal Inference
Czasopisma -
Energy-Efficient Neural Network Inference with Microcavity Exciton Polaritons
Publikacja -
Inference Mechanisms for Knowledge Management System in E-health Environment
PublikacjaW artykule zaprezentowano badania i ich wyniki osiągnięte w trakcie prac na systemem zarządzania wiedzą dla systemu typu e-health PIPS (ang. Personalised Information Platform for Life and Health Services)Opisano warstwę semantyczną systemu zarządzania wiedzą, a w szczególności skoncentrowano się na silniku wnioskującym.Poddano dyskusji następujące zagadnienia: analizę dostępnych narzędzi wnioskujący i ich zgodności ze standardem...
-
Estimation of Housing Demand with Adaptive Neuro-Fuzzy Inference Systems (ANFIS)
PublikacjaIt has always been important to anticipate the demand for a product. To determine the demand for any product, the parameters such as the economic situation and the demands of the rival products are used generally. Especially in the housing sector, which is the locomotive sector for emerging countries, it is critical to anticipate housing demand and its relationship with economic variables. Because of that, economists, real estate...
-
Estimation of Housing Demand with Adaptive Neuro-Fuzzy Inference Systems (ANFIS)
Publikacja -
Information and Inference-A Journal of the IMA
Czasopisma -
Statistical Inference for Stochastic Processes
Czasopisma -
JOURNAL OF STATISTICAL PLANNING AND INFERENCE
Czasopisma -
Optical Sensor Based Gestures Inference Using Recurrent Neural Network in Mobile Conditions
PublikacjaIn this paper the implementation of recurrent neural network models for hand gesture recognition on edge devices was performed. The models were trained with 27 hand gestures recorded with the use of a linear optical sensor consisting of 8 photodiodes and 4 LEDs. Different models, trained off-line, were tested in terms of different network topologies (different number of neurons and layers) and different effective sampling frequency...
-
Effect of adopted rules of inference and methods of defuzzification on the final result of the evaluation of reliability made using the fuzzy logic methods
PublikacjaThe object of interest is to solve the problem of risk management of marine systems. But the main trouble is a lack of numerous and sure data on the reliability of the components of such systems. The methods based on the fuzzy logic seem to be helpful here. The goal of the article is to check the effect of using different fuzzy inference rules and methods of defuzzification on the final result of reliability assessment. The three...
-
Multi-objective optimization of tool wear, surface roughness, and material removal rate in finishing honing processes using adaptive neural fuzzy inference systems
PublikacjaHoning processes are usually employed to manufacture combustion engine cylinders and hydraulic cylinders. A crosshatch pattern is obtained that favors the oil flow. In this paper, Adaptive Neural Fuzzy Inference System (ANFIS) models were obtained for tool wear, average roughness Ra, cylindricity and material removal rate in finish honing processes. In addition, multi-objective optimization with the desirability function method...
-
Piotr Szczuko dr hab. inż.
OsobyDr hab. inż. Piotr Szczuko w 2002 roku ukończył studia na Wydziale Elektroniki, Telekomunikacji i Informatyki Politechniki Gdańskiej zdobywając tytuł magistra inżyniera. Tematem pracy dyplomowej było badanie zjawisk jednoczesnej percepcji obrazu cyfrowego i dźwięku dookólnego. W roku 2008 obronił rozprawę doktorską zatytułowaną "Zastosowanie reguł rozmytych w komputerowej animacji postaci", za którą otrzymał nagrodę Prezesa Rady...
-
Reliability and types of diagnosis in the process of diesel engine operation
PublikacjaThe article presents complexity of the problem concerning development of diagnosis with defined reliability by a diagnosing system (SDG) on technical condition of marine combustion engines, especially main engines. It was shown that development of the final diagnosis, the so-called initial operation diagnosis, on the operational usability (fDG on PEx) of a main engine in particular, is not possible without prior development of...
-
Performance Analysis of Convolutional Neural Networks on Embedded Systems
PublikacjaMachine learning is no longer confined to cloud and high-end server systems and has been successfully deployed on devices that are part of Internet of Things. This paper presents the analysis of performance of convolutional neural networks deployed on an ARM microcontroller. Inference time is measured for different core frequencies, with and without DSP instructions and disabled access to cache. Networks use both real-valued and...
-
Syntactic modular decomposition of large ontologies with relational database
PublikacjaSupport for modularity allows complex ontologies to be separated into smaller pieces (modules) that are easier to maintain and compute. Instead of considering the entire complex ontology, users may benefit more by starting from a problem-specific set of concepts (signature of problem) from the ontology and exploring its surrounding logical modules. Additionally, an ontology modularization mechanism allows for the splitting up of...
-
Fuzzy rule-based dynamic gesture recognition employing camera & multimedia projector
PublikacjaIn the paper the system based on camera and multimedia projector enabling a user to control computer applications by dynamic hand gestures is presented. The main objective is to present the gesture recognition methodology which bases on representing hand movement trajectory by motion vectors analyzed using fuzzy rule-based inference. The approach was engineered in the system developed with J2SE and C++ / OpenCV technology. OpenCV...
-
Deep Learning Optimization for Edge Devices: Analysis of Training Quantization Parameters
PublikacjaThis paper focuses on convolution neural network quantization problem. The quantization has a distinct stage of data conversion from floating-point into integer-point numbers. In general, the process of quantization is associated with the reduction of the matrix dimension via limited precision of the numbers. However, the training and inference stages of deep learning neural network are limited by the space of the memory and a...
-
Food Classification from Images Using a Neural Network Based Approach with NVIDIA Volta and Pascal GPUs
PublikacjaIn the paper we investigate the problem of food classification from images, for the Food-101 dataset extended with 31 additional food classes from Polish cuisine. We adopted transfer learning and firstly measured training times for models such as MobileNet, MobileNetV2, ResNet50, ResNet50V2, ResNet101, ResNet101V2, InceptionV3, InceptionResNetV2, Xception, NasNetMobile and DenseNet, for systems with NVIDIA Tesla V100 (Volta) and...
-
Utilization of fuzzy rules in computer character animation
PublikacjaThe chapter presents a method for automatic enhancement of computer character animation utilizing fuzzy inference. First the user designs a prototype version of animation, with keyframes only for important poses, roughly describing the action. Then animation is enriched with new motion phases calculated by the fuzzy inference system using descriptors given by the user. Various degrees of motion fluency and naturalness are possible...
-
Detection of anomalies in bee colony using transitioning state and contrastive autoencoders
PublikacjaHoneybees plays vital role for the environmental sustainability and overall agricultural economy. Assisting bee colonies within their proper functioning brings the attention of researchers around the world. Electronics systems and machine learning algorithms are being developed for classifying specific undesirable bee behaviors in order to alert about upcoming substantial losses. However, classifiers could be impaired when used...
-
Unveiling the Pool of Metallophores in Native Environments and Correlation with Their Potential Producers
PublikacjaFor many organisms, metallophores are essential biogenic ligands that ensure metal scavenging and acquisition from their environment. Their identification is challenging in highly organic matter rich environments like peatlands due to low solubilization and metal scarcity and high matrix complexity. In contrast to common approaches based on sample modification by spiking of metal isotope tags, we have developed a two-dimensional...
-
The impact of technological and structural changes in the national economy on the labour-capital relations
PublikacjaThe aim of the research presented in this paper is to present the relations between labour and capital in the national economy, resulting from technological and structural changes taking place in the years 1991 to 2008. The structure of the paper is as follows. It first presents the functional determinants for the Polish economy in 1991-2008 affecting the phenomenon subject to study. Then it presents the preliminary analysis of...
-
Semantic technologies based method of collection, processing and sharing information along food chain
PublikacjaIn the paper the method of collecting, processing and sharing information along food chain is presented. Innovative features of that method result from advantages of data engineering based on semantic technologies. The source to build ontology are standards and regulations related to food production, and data collected in databases owned by food chain participants. It allows food chain information resources can be represented in...
-
Diagnostic tolerances' evaluation method of the start-up exhaust temperature of a naval gas turbine
PublikacjaThe conducted investigations aimed to elaborate the method of marking diagnostic tolerances of the exhaust temperature of a gas turbine observed during engine's start-up process. The diagnostic tolerances were determined by means of statistical inference by creating the hypothesis about a normal distribution of the start-up exhaust temperature's dispersion in the initial operation moment, which was subsequently verified applying...
-
Expert assessment of arguments: a method and its experimental evaluation
PublikacjaArgument structures are commonly used to develop and present cases for safety, security and other properties. Such argument structures tend to grow excessively. To deal with this problem, appropriate methods of their assessment are required. Two objectives are of particular interest: (1) systematic and explicit assessment of the compelling power of an argument, and (2) communication of the result of such an assessment to relevant...
-
Survey on fuzzy logic methods in control systems of electromechanical plants
PublikacjaРассмотрены алгоритмы управления электромеханическими системами с использованием теории нечеткой логики, приводятся основные положения их синтеза, рассматриваются методы анализа их устойчивости на основе нечетких функций Ляпунова. Эти алгоритмы чаще всего реализуются в виде различных регуляторов, применение которых целесообразно в системах, математическая модель которых не известна, не детерминирована или является строго нелинейной,...
-
Influence of lubricating oil improvers on performance of crankshaft seals
PublikacjaThis paper presents an original method for checking influence of lubricating oil improvers on performance of crankshaft seals of combustion piston engine. Crankshaft seals were tested with the use of a modified friction node of T-02 four- ball apparatus in laboratory conditions. The tests were conducted according to a worked-out algorithm. Their results confirmed usefulness of the method for determining „harmful” performance...
-
New method for personalization of avatar animation
PublikacjaThe paper presents a method for creating a personalized animation of avatar utilizing fuzzy inference. First the user designs a prototype version of animation, with keyframes only for important poses, roughly describing the action. Then animation is enriched with new motion phases calculated by the fuzzy inference system using descriptors given by the user. Various degrees of motion fluency and naturalness are possible to achieve....
-
Optymalizacja treningu i wnioskowania sieci neuronowych
PublikacjaSieci neuronowe są jedną z najpopularniejszych i najszybciej rozwijających się dziedzin sztucznej inteligencji. Ich praktyczne wykorzystanie umożliwiło szersze użycie komputerów w wielu obszarach komunikacji, przemysłu i transportu. Dowody tego są widoczne w elektronice użytkowej, medycynie, a nawet w zastosowaniach militarnych. Wykorzystanie sztucznej inteligencji w wielu przypadkach wymaga jednak znacznej mocy obliczeniowej,...
-
Hand gesture recognition supported by fuzzy rules and Kalman filters
PublikacjaThe paper presents a system based on camera and multimediaprojector enabling a user to control computer applications by dynamic hand gestures. Gesture recognition methodology based on representing hand movement trajectory by motion vectors analysed using fuzzy rule-based inference is first given. For effective hand position tracking Kalman filters are employed. The system engineered is developed using J2SE and C++/OpenCV technology....
-
Computer animation system based on rough sets and fuzzy logic
PublikacjaA fuzzy logic inference system was created, based on the analysis of animated motion features. The objective of the system is to facilitate the creation of high quality animation by analyzing personalized styles contained in numerous animations. Sequences portraying a virtual character acting with a differentiating personalized style (natural or exaggerated) and various levels of fluidity were prepared and subjectively evaluated....
-
Residual MobileNets
PublikacjaAs modern convolutional neural networks become increasingly deeper, they also become slower and require high computational resources beyond the capabilities of many mobile and embedded platforms. To address this challenge, much of the recent research has focused on reducing the model size and computational complexity. In this paper, we propose a novel residual depth-separable convolution block, which is an improvement of the basic...
-
PROJECT OF AN TDC SIGNAL DISTRIBUTOR PROTOTYPE WITH GAL-VANIC SEPARATION DESTINED FOR STUDIES OF THE BEGINNING OF SELFIGNITION USING LANGMUIR PROBE
PublikacjaThe knowledge of pistons' TDC is crucial while performing any kind of measurement beneficial to diagnostic inference of a diesel engine. The research of engines under exploitation often cause impartial difficulties with, eg. capability of probe installation. Therefore, it becomes crucial to minimalize those. The paper presents a description of a TDC impulse distributor - divider that provides signal delivery to more than one measurement...
-
Exhaust gas temperature measurements in diagnostic examination of naval gas turbine engines. Part III. Diagnostic and operating tolerances
PublikacjaThe third part of the article presents a method for detecting failures of the automatic engine control system with the aid of an exhaust gas temperature setter, specially designed and machined for this purpose. It also presents a procedure of identifying the operating tolerances and determining the diagnostic tolerances for the exhaust gas temperature recorded in the naval turbine engine during the start-up and acceleration processes....
-
Cognitum Ontorion: Knowledge Representation and Reasoning System
PublikacjaAt any point of human activity, knowledge and expertise are a key factors in understanding and solving any given problem. In present days, computer systems have the ability to support their users in an efficient and reliable way in gathering and processing knowledge. In this chapter we show how to use Cognitum Ontorion system in this areas. In first section, we identify emerging issues focused on how to represent and inference...
-
Visual Features for Improving Endoscopic Bleeding Detection Using Convolutional Neural Networks
PublikacjaThe presented paper investigates the problem of endoscopic bleeding detection in endoscopic videos in the form of a binary image classification task. A set of definitions of high-level visual features of endoscopic bleeding is introduced, which incorporates domain knowledge from the field. The high-level features are coupled with respective feature descriptors, enabling automatic capture of the features using image processing methods....
-
Conception for diagnosing SI engines fed with biofuel in operation conditions
PublikacjaThe article deals with the question of diagnosing the high power self ignition (SI) engines fed with pro-ecologic fuels in an aspect of appearing operation problems. There has been presented the specific of action and techno-operation profile of the combustion engine foreseen as an investigations object. The preliminary conception of diagnostic investigations of the engine 7L35MC type produced in H. Cegielski - Poznan S.A. factory...
-
A probabilistic-driven framework for enhanced corrosion estimation of ship structural components
PublikacjaThe work proposes a probabilistic-driven framework for enhanced corrosion estimation of ship structural components using Bayesian inference and limited measurement data. The new approach for modelling measurement uncertainty is proposed based on the results of previous corrosion tests that incorporate the non-uniform character of the corroded surface of structural components. The proposed framework's basic features are outlined,...
-
Characterizing the Scalability of Graph Convolutional Networks on Intel® PIUMA
PublikacjaLarge-scale Graph Convolutional Network (GCN) inference on traditional CPU/GPU systems is challenging due to a large memory footprint, sparse computational patterns, and irregular memory accesses with poor locality. Intel’s Programmable Integrated Unffied Memory Architecture (PIUMA) is designed to address these challenges for graph analytics. In this paper, a detailed characterization of GCNs is presented using the Open-Graph Benchmark...
-
On reliability assessment of ship machinery system in different autonomy degree; A Bayesian-based approach
PublikacjaAnalyzing the reliability of autonomous ships has recently attracted attention mainly due to epistemic uncertainty (lack of knowledge) integrated with automatic operations in the maritime sector. The advent of new random failures with unrecognized failure patterns in autonomous ship operations requires a comprehensive reliability assessment specifically aiming at estimating the time in which the ship can be trusted to be left unattended....
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublikacjaRecent years have seen impressive progress in visual recognition on many benchmarks, however, generalization to the out-of-distribution setting remains a significant challenge. A state-of-the-art method for robust visual recognition is model ensembling. However, recently it was shown that similarly competitive results could be achieved with a much smaller cost, by using multi-input multi-output architecture (MIMO). In this work,...
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublikacjaRecent years have seen impressive progress in visual recognition on many benchmarks, however, generalization to the out-of-distribution setting remains a significant challenge. A state-of-the-art method for robust visual recognition is model ensembling. However, recently it was shown that similarly competitive results could be achieved with a much smaller cost, by using multi-input multi-output architecture (MIMO). In this work,...
-
Speaker Recognition Using Convolutional Neural Network with Minimal Training Data for Smart Home Solutions
PublikacjaWith the technology advancements in smart home sector, voice control and automation are key components that can make a real difference in people's lives. The voice recognition technology market continues to involve rapidly as almost all smart home devices are providing speaker recognition capability today. However, most of them provide cloud-based solutions or use very deep Neural Networks for speaker recognition task, which are...
-
Face with Mask Detection in Thermal Images Using Deep Neural Networks
PublikacjaAs the interest in facial detection grows, especially during a pandemic, solutions are sought that will be effective and bring more benefits. This is the case with the use of thermal imaging, which is resistant to environmental factors and makes it possible, for example, to determine the temperature based on the detected face, which brings new perspectives and opportunities to use such an approach for health control purposes. The...
-
DNA methylation as a mediator of HLA-DRB1*15:01 and a protective variant in multiple sclerosis
PublikacjaThe human leukocyte antigen (HLA) haplotype DRB1*15:01 is the major risk factor for multiple sclerosis (MS). Here, we find that DRB1*15:01 is hypomethylated and predominantly expressed in monocytes among carriers of DRB1*15:01. A differentially methylated region (DMR) encompassing HLA-DRB1 exon 2 is particularly affected and displays methylation-sensitive regulatory properties in vitro. Causal inference and Mendelian randomization...
-
Design of three control algorithms for an averaging tank with variable filing
PublikacjaAn averaging tank with variable filling is a nonlinear multidimensional system and can thus be considered a complex control sys-tem. General control objectives of such object include ensuring stability, zero steady state error and achieving simultaneously shortest possible settling time and minimal overshoot. The main purpose of this research work was the modelling and synthesis of three control systems for an averaging tank. In...
-
Enhancement of computer character animation utilizing fuzzy rules
PublikacjaRozdział przedstawia nową metodę przetwarzania komputerowych animacji postaci. Wykorzystuje ona wnioskowanie rozmyte, oparte na regułach i funkcjach przynależności uzyskanych w procesie analizy wyników testów subiektywnej oceny jakości animacji. W trakcie przetwarzania do animacji automatycznie dodawane są nowe fazy ruchu, co skutkuje poprawą jakości wizualnej oraz zmianą płynności i stylizacji ruchu w sposób zamierzony. W referacie...
-
A Computationally Efficient Model for Predicting Successful Memory Encoding Using Machine-Learning-based EEG Channel Selection
PublikacjaComputational cost is an important consideration for memory encoding prediction models that use data from dozens of implanted electrodes. We propose a method to reduce computational expense by selecting a subset of all the electrodes to build the prediction model. The electrodes were selected based on their likelihood of measuring brain activity useful for predicting memory encoding better than chance (in terms of AUC). A logistic...
-
Laxer Clinical Criteria for Gaming Disorder May Hinder Future Efforts to Devise an Efficient Diagnostic Approach: A Tree-Based Model Study
PublikacjaInternet Gaming Disorder (IGD) has been recognized in May 2013 and can be evaluated using the criteria developed by American Psychiatric Association (APA). The present study investigated the role each IGD criteria plays in diagnosing disordered gaming. A total of 3,377 participants (mean age 20 years, SD = 4.3 years) participated in the study. The data collected was scrutinized to detect patterns...