Vibration Detection of Vehicle Impact Using Smartphone Accelerometer Data and Long-Short Term Memory Neural Network

2020 ◽  
Author(s):  
João Paulo Brognoni Casati ◽  
Ruy Alberto Corrêa Altafim ◽  
Ruy Alberto Pisani Altafim
Keyword(s):  
Neural Network ◽  
Short Term Memory ◽  
Smart Cities ◽  
Accelerometer Data ◽  
Vehicle Type ◽  
Vehicle Impact ◽  
Impact Vibration ◽  
Traffic Behavior ◽  
Vehicle Type Classification

Vehicle's analysis can be useful for a variety of traffic problems, such as monitoring road damages and vehicle type classification. Further, traffic behavior analysis can be useful to monitor traffic jams as a smart cities solution. In this paper, the vibration caused by a vehicle passing through a speed bump was recorded with a docked smartphone. The acquired signals were processed in order to detect the generated impact. In order to analyze this data a LSTM neural network was used due to its classification process over time while the smartphone accelerometer was continuously operating (waiting for a vehicle pass by). This deep learning technique allows the use of raw 3-axis accelerometer data. The results achieved 98% of accuracy with a low level of false positives (less than 1%). Indicating that the methodology is effective in classification of vehicles by their impact vibration.

scholarly journals Classification of Analyzed Text in Speech Recognition Using RNN-LSTM in Comparison with Convolutional Neural Network to Improve Precision for Identification of Keywords

2021 ◽  
Vol 11 (2) ◽  
pp. 1097-1108
Author(s):  
Bathaloori Reddy Prasad
Keyword(s):  
Neural Network ◽  
Speech Recognition ◽  
Convolutional Neural Network ◽  
Recurrent Neural Network ◽  
Short Term Memory ◽  
Statistical Significance ◽  
Language Translation ◽  
Accuracy And Precision ◽  
Long Short Term Memory

Aim: Text classification is a method to classify the features from language translation in speech recognition from English to Telugu using a recurrent neural network- long short term memory (RNN-LSTM) comparison with convolutional neural network (CNN). Materials and Methods: Accuracy and precision are performed with dataset alexa and english-telugu of size 8166 sentences. Classification of language translation is performed by the recurrent neural network where a number of the samples (N=62) and convolutional neural network were a number of samples (N=62) techniques, the algorithm RNN implies speech recognition that can be compared with convolutional is the second technique. Results and Discussion: RNN-LSTM from the dataset speech recognition, feature Telugu_id produce accuracy 93% and precision 68.04% which can be comparatively higher than CNN accuracy 66.11%, precision 61.90%. It shows a statistical significance as 0.007 from Independent Sample T-test. Conclusion: The RNN-LSTM performs better in finding accuracy and precision when compared to CNN.

scholarly journals Comparison of machine learning and deep learning techniques in promoter prediction across diverse species

2021 ◽  
Vol 7 ◽  
pp. e365
Author(s):  
Nikita Bhandari ◽  
Satyajeet Khare ◽  
Rahee Walambe ◽  
Ketan Kotecha
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Short Term Memory ◽  
Regulatory Elements ◽  
Gene Promoters ◽  
Promoter Prediction ◽  
Promoter Sequences ◽  
Learning Techniques

Gene promoters are the key DNA regulatory elements positioned around the transcription start sites and are responsible for regulating gene transcription process. Various alignment-based, signal-based and content-based approaches are reported for the prediction of promoters. However, since all promoter sequences do not show explicit features, the prediction performance of these techniques is poor. Therefore, many machine learning and deep learning models have been proposed for promoter prediction. In this work, we studied methods for vector encoding and promoter classification using genome sequences of three distinct higher eukaryotes viz. yeast (Saccharomyces cerevisiae), A. thaliana (plant) and human (Homo sapiens). We compared one-hot vector encoding method with frequency-based tokenization (FBT) for data pre-processing on 1-D Convolutional Neural Network (CNN) model. We found that FBT gives a shorter input dimension reducing the training time without affecting the sensitivity and specificity of classification. We employed the deep learning techniques, mainly CNN and recurrent neural network with Long Short Term Memory (LSTM) and random forest (RF) classifier for promoter classification at k-mer sizes of 2, 4 and 8. We found CNN to be superior in classification of promoters from non-promoter sequences (binary classification) as well as species-specific classification of promoter sequences (multiclass classification). In summary, the contribution of this work lies in the use of synthetic shuffled negative dataset and frequency-based tokenization for pre-processing. This study provides a comprehensive and generic framework for classification tasks in genomic applications and can be extended to various classification problems.

A comparative analysis of euphemistic sentences in news using feature weight scheme and intelligent techniques

2021 ◽  
pp. 1-12
Author(s):  
K. Seethappan ◽  
K. Premalatha
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Short Term Memory ◽  
Automatic Classification ◽  
Weighting Scheme ◽  
Feature Weighting ◽  
Classification Algorithms ◽  
Weighting Schemes

Although there have been various researches in the detection of different figurative language, there is no single work in the automatic classification of euphemisms. Our primary work is to present a system for the automatic classification of euphemistic phrases in a document. In this research, a large dataset consisting of 100,000 sentences is collected from different resources for identifying euphemism or non-euphemism utterances. In this work, several approaches are focused to improve the euphemism classification: 1. A Combination of lexical n-gram features 2.Three Feature-weighting schemes 3.Deep learning classification algorithms. In this paper, four machine learning (J48, Random Forest, Multinomial Naïve Bayes, and SVM) and three deep learning algorithms (Multilayer Perceptron, Convolutional Neural Network, and Long Short-Term Memory) are investigated with various combinations of features and feature weighting schemes to classify the sentences. According to our experiments, Convolutional Neural Network (CNN) achieves precision 95.43%, recall 95.06%, F-Score 95.25%, accuracy 95.26%, and Kappa 0.905 by using a combination of unigram and bigram features with TF-IDF feature weighting scheme in the classification of euphemism. These results of experiments show CNN with a strong combination of unigram and bigram features set with TF-IDF feature weighting scheme outperforms another six classification algorithms in detecting the euphemisms in our dataset.

scholarly journals Sentimental Classification of News Headlines using Recurrent Neural Network

2020 ◽  
Vol 9 (6) ◽  
pp. 207-210
Keyword(s):  
Neural Network ◽  
Language Processing ◽  
Recurrent Neural Network ◽  
Short Term Memory ◽  
Attention Mechanism ◽  
Machine Learning Techniques ◽  
Learning Techniques ◽  
Negative Comments ◽  
News Headlines

Sentiment analysis combines the natural language processing task and analysis of the text that attempts to predict the sentiment of the text in terms of positive and negative comments. Nowadays, the tremendous volume of news originated via different webpages, and it is feasible to determine the opinion of particular news. This work tries to judge completely various machine learning techniques to classify the view of the news headlines. In this project, propose the appliance of Recurrent Neural Network with Long Short Term Memory Unit(LSTM), focus on seeking out similar news headlines, and predict the opinion of news headlines from numerous sources. The main objective is to classify the sentiment of news headlines from various sources using a recurrent neural network. Interestingly, the proposed attention mechanism performs better than the more complex attention mechanism on a held-out set of articles.

scholarly journals CLASSIFICATION OF DEMENTIA DISEASES USING DEEP LEARNING TECHNIQUES

2020 ◽  
Vol 4 (2) ◽  
pp. 371-379
Author(s):  
David.O. Oyewola ◽  
Bernard Alechenu ◽  
Kuluwa A. Al-Mustapha ◽  
Oluwatoyosi .V. Oyewande
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Early Diagnosis ◽  
Short Term Memory ◽  
Economic Status ◽  
Generalized Regression Neural Network ◽  
Intracranial Volume ◽  
Clinical Dementia Rating ◽  
Learning Techniques

Dementia is the most frequent degenerative sickness in adults where early diagnosis can forestall or prolong progression. In this study, we used a deep learning techniques for classification of dementia. Data were collected from OASIS database of all the patients receiving dementia screening. The data included the patient’s sex, age, education, social economic status, Mini-Mental State Examination, Clinical Dementia Rating, Atlas Scaling Factor, Estimated Total Intracranial Volume and Normalized Whole Brain Volume. The performance of every algorithm is juxtaposed with Generalized Regression Neural Network (GRNN), Radial Basis Neural Network (RBNN), Multilayer Perceptron Neural Network (MPNN) and Long Short Term Memory (LSTM) using Sensitivity, Specificity, Detection Rate. The results show that with 100% efficiency, GRNN, RBNN and LSTM tend to be the best in the classification of dementia. The use of deep learning such as LSTM for early diagnosis of dementia can help improve the process of dementia diagnosis.

Sign in / Sign up

Export Citation Format

Share Document