Search results for "feature"
showing 10 items of 4091 documents
Adaptive Continuous Feature Binarization for Tsetlin Machines Applied to Forecasting Dengue Incidences in the Philippines
2020
The Tsetlin Machine (TM) is a recent interpretable machine learning algorithm that requires relatively modest computational power, yet attains competitive accuracy in several benchmarks. TMs are inherently binary; however, many machine learning problems are continuous. While binarization of continuous data through brute-force thresholding has yielded promising accuracy, such an approach is computationally expensive and hinders extrapolation. In this paper, we address these limitations by standardizing features to support scale shifts in the transition from training data to real-world operation, typical for e.g. forecasting. For scalability, we employ sampling to reduce the number of binariz…
A Review of Kernel Methods in Remote Sensing Data Analysis
2011
Kernel methods have proven effective in the analysis of images of the Earth acquired by airborne and satellite sensors. Kernel methods provide a consistent and well-founded theoretical framework for developing nonlinear techniques and have useful properties when dealing with low number of (potentially high dimensional) training samples, the presence of heterogenous multimodalities, and different noise sources in the data. These properties are particularly appropriate for remote sensing data analysis. In fact, kernel methods have improved results of parametric linear methods and neural networks in applications such as natural resource control, detection and monitoring of anthropic infrastruc…
State classification for autonomous gas sample taking using deep convolutional neural networks
2017
Despite recent rapid advances and successful large-scale application of deep Convolutional Neural Networks (CNNs) using image, video, sound, text and time-series data, its adoption within the oil and gas industry in particular have been sparse. In this paper, we initially present an overview of opportunities for deep CNN methods within oil and gas industry, followed by details on a novel development where deep CNN have been used for state classification of autonomous gas sample taking procedure utilizing an industrial robot. The experimental results — using a deep CNN containing six layers — show accuracy levels exceeding 99 %. In addition, the advantages of using parallel computing with GP…
CostNet: An End-to-End Framework for Goal-Directed Reinforcement Learning
2020
Reinforcement Learning (RL) is a general framework concerned with an agent that seeks to maximize rewards in an environment. The learning typically happens through trial and error using explorative methods, such as \(\epsilon \)-greedy. There are two approaches, model-based and model-free reinforcement learning, that show concrete results in several disciplines. Model-based RL learns a model of the environment for learning the policy while model-free approaches are fully explorative and exploitative without considering the underlying environment dynamics. Model-free RL works conceptually well in simulated environments, and empirical evidence suggests that trial and error lead to a near-opti…
Ventricular fibrillation detection from ECG surface electrodes using different filtering techniques, window length and artificial neural networks
2017
Medical personnel face many difficulties when diagnosing ventricular fibrillation (VF). Its correct diagnosis allows to decide the right medical treatment and, therefore, it is essential to tell it apart adequately from ventricular tachycardia (VT) and other arrhythmias. If the required therapy is not appropriate, the personnel could cause serious injuries or even induce VF. In this work, a diagnosis automatic system for the detection of VF through feature extraction was developed. To verify the validity of this method, an Artificial Neural Network (ANN) classifier was used. The ECG signals used were obtained from the MIT-BIH Malignant Ventricular Arrhythmia Database and AHA (2000 series) d…
Combining Auto-Encoder with LSTM for WiFi-Based Fingerprint Positioning
2021
Although indoor positioning has long been investigated by various means, its accuracy remains concern. Several recent studies have applied machine learning algorithms to explore wireless fidelity (WiFi)-based positioning. In this paper, we propose a novel deep learning model which concatenates an auto-encoder with a long short term memory (LSTM) network for the purpose of WiFi fingerprint positioning. We first employ an auto-encoder to extract representative latent codes of fingerprints. Such an extraction is proven to be more reliable than simply using a deep neural network to extract representative features since a latent code can be reverted back to its original input. Then, a sequence o…
Connectionist models of face processing: A survey
1994
Abstract Connectionist models of face recognition, identification, and categorization have appeared recently in several disciplines, including psychology, computer science, and engineering. We present a review of these models with the goal of complementing a recent survey by Samal and Iyengar [Pattern Recognition25, 65–77 (1992)] of nonconnectionist approaches to the problem of the automatic face recognition. We concentrate on models that use linear autoassociative networks, nonlinear autoassociative (or compression) and/or heteroassociative backpropagation networks. One advantage of these models over some nonconnectionist approaches is that analyzable features emerge naturally from image-b…
Logo detection in images using HOG and SIFT
2017
In this paper we present a study of logo detection in images from a media agency. We compare two most widely used methods — HOG and SIFT on a challenging dataset of images arising from a printed press and news portals. Despite common opinion that SIFT method is superior, our results show that HOG method performs significantly better on our dataset. We augment the HOG method with image resizing and rotation to improve its performance even more. We found out that by using such approach it is possible to obtain good results with increased recall and reasonably decreased precision.
An Encrypted Traffic Classification Framework Based on Convolutional Neural Networks and Stacked Autoencoders
2020
In recent years, deep learning-based encrypted traffic classification has proven to be effective; especially, using neural networks to extract features from raw traffic to classify encrypted traffic. However, most of the neural networks need a fixed-sized input, so that the raw traffic need to be trimmed. This will cause the loss of some information; for example, we do not know the number of packets in a session. To solve these problems, a framework, which implements both a convolutional neural network (CNN) and a stacked autoencoder (SAE), is proposed in this paper. This framework uses a CNN to extract high-level features from raw network traffic and uses an SAE to encode the 26 statistica…
Présence d’Asellides stygobies (Crustacea, Isopoda, Aselloidea) dans la région du Primorye, Sibérie sud-orientale
1993
Two species of anophthalmous, unpigmentedasellids have been found in springs and groundwaters of S.E. Siberia (Primorye region). Asellus (Asellus) primoryensis n. sp. is closelyrelated to the epigean species A. (A.) hilgendorfii Bovallius, 1886, as is the case for all stygobiont Asellus (Asellus) species previously known from the Japanese archipelago. Sibirasellus parpurae n. g., n. sp. is closely related to the microphthalmous species Asellus dentifer Birstein & Levanidov, 1952 from the Ussuri Basin (Khor region), now type-species of the new genus Sibirasellus. These two species show several original characters: body covered by numerous cuticular squamulae, mandibular palp reduced (gla…