Search results for: label noise
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 1393

Search results for: label noise

1393 The Influence of Noise on Aerial Image Semantic Segmentation

Authors: Pengchao Wei, Xiangzhong Fang

Abstract:

Noise is ubiquitous in this world. Denoising is an essential technology, especially in image semantic segmentation, where noises are generally categorized into two main types i.e. feature noise and label noise. The main focus of this paper is aiming at modeling label noise, investigating the behaviors of different types of label noise on image semantic segmentation tasks using K-Nearest-Neighbor and Convolutional Neural Network classifier. The performance without label noise and with is evaluated and illustrated in this paper. In addition to that, the influence of feature noise on the image semantic segmentation task is researched as well and a feature noise reduction method is applied to mitigate its influence in the learning procedure.

Keywords: convolutional neural network, denoising, feature noise, image semantic segmentation, k-nearest-neighbor, label noise

Procedia PDF Downloads 219
1392 Deep learning with Noisy Labels : Learning True Labels as Discrete Latent Variable

Authors: Azeddine El-Hassouny, Chandrashekhar Meshram, Geraldin Nanfack

Abstract:

In recent years, learning from data with noisy labels (Label Noise) has been a major concern in supervised learning. This problem has become even more worrying in Deep Learning, where the generalization capabilities have been questioned lately. Indeed, deep learning requires a large amount of data that is generally collected by search engines, which frequently return data with unreliable labels. In this paper, we investigate the Label Noise in Deep Learning using variational inference. Our contributions are : (1) exploiting Label Noise concept where the true labels are learnt using reparameterization variational inference, while observed labels are learnt discriminatively. (2) the noise transition matrix is learnt during the training without any particular process, neither heuristic nor preliminary phases. The theoretical results shows how true label distribution can be learned by variational inference in any discriminate neural network, and the effectiveness of our approach is proved in several target datasets, such as MNIST and CIFAR32.

Keywords: label noise, deep learning, discrete latent variable, variational inference, MNIST, CIFAR32

Procedia PDF Downloads 126
1391 Learning from Small Amount of Medical Data with Noisy Labels: A Meta-Learning Approach

Authors: Gorkem Algan, Ilkay Ulusoy, Saban Gonul, Banu Turgut, Berker Bakbak

Abstract:

Computer vision systems recently made a big leap thanks to deep neural networks. However, these systems require correctly labeled large datasets in order to be trained properly, which is very difficult to obtain for medical applications. Two main reasons for label noise in medical applications are the high complexity of the data and conflicting opinions of experts. Moreover, medical imaging datasets are commonly tiny, which makes each data very important in learning. As a result, if not handled properly, label noise significantly degrades the performance. Therefore, a label-noise-robust learning algorithm that makes use of the meta-learning paradigm is proposed in this article. The proposed solution is tested on retinopathy of prematurity (ROP) dataset with a very high label noise of 68%. Results show that the proposed algorithm significantly improves the classification algorithm's performance in the presence of noisy labels.

Keywords: deep learning, label noise, robust learning, meta-learning, retinopathy of prematurity

Procedia PDF Downloads 160
1390 A New Approach for Improving Accuracy of Multi Label Stream Data

Authors: Kunal Shah, Swati Patel

Abstract:

Many real world problems involve data which can be considered as multi-label data streams. Efficient methods exist for multi-label classification in non streaming scenarios. However, learning in evolving streaming scenarios is more challenging, as the learners must be able to adapt to change using limited time and memory. Classification is used to predict class of unseen instance as accurate as possible. Multi label classification is a variant of single label classification where set of labels associated with single instance. Multi label classification is used by modern applications, such as text classification, functional genomics, image classification, music categorization etc. This paper introduces the task of multi-label classification, methods for multi-label classification and evolution measure for multi-label classification. Also, comparative analysis of multi label classification methods on the basis of theoretical study, and then on the basis of simulation was done on various data sets.

Keywords: binary relevance, concept drift, data stream mining, MLSC, multiple window with buffer

Procedia PDF Downloads 583
1389 An Improvement of Multi-Label Image Classification Method Based on Histogram of Oriented Gradient

Authors: Ziad Abdallah, Mohamad Oueidat, Ali El-Zaart

Abstract:

Image Multi-label Classification (IMC) assigns a label or a set of labels to an image. The big demand for image annotation and archiving in the web attracts the researchers to develop many algorithms for this application domain. The existing techniques for IMC have two drawbacks: The description of the elementary characteristics from the image and the correlation between labels are not taken into account. In this paper, we present an algorithm (MIML-HOGLPP), which simultaneously handles these limitations. The algorithm uses the histogram of gradients as feature descriptor. It applies the Label Priority Power-set as multi-label transformation to solve the problem of label correlation. The experiment shows that the results of MIML-HOGLPP are better in terms of some of the evaluation metrics comparing with the two existing techniques.

Keywords: data mining, information retrieval system, multi-label, problem transformation, histogram of gradients

Procedia PDF Downloads 373
1388 Effective Nutrition Label Use on Smartphones

Authors: Vladimir Kulyukin, Tanwir Zaman, Sarat Kiran Andhavarapu

Abstract:

Research on nutrition label use identifies four factors that impede comprehension and retention of nutrition information by consumers: label’s location on the package, presentation of information within the label, label’s surface size, and surrounding visual clutter. In this paper, a system is presented that makes nutrition label use more effective for nutrition information comprehension and retention. The system’s front end is a smartphone application. The system’s back end is a four node Linux cluster for image recognition and data storage. Image frames captured on the smartphone are sent to the back end for skewed or aligned barcode recognition. When barcodes are recognized, corresponding nutrition labels are retrieved from a cloud database and presented to the user on the smartphone’s touchscreen. Each displayed nutrition label is positioned centrally on the touchscreen with no surrounding visual clutter. Wikipedia links to important nutrition terms are embedded to improve comprehension and retention of nutrition information. Standard touch gestures (e.g., zoom in/out) available on mainstream smartphones are used to manipulate the label’s surface size. The nutrition label database currently includes 200,000 nutrition labels compiled from public web sites by a custom crawler. Stress test experiments with the node cluster are presented. Implications for proactive nutrition management and food policy are discussed.

Keywords: mobile computing, cloud computing, nutrition label use, nutrition management, barcode scanning

Procedia PDF Downloads 372
1387 A Paradigm for Characterization and Checking of a Human Noise Behavior

Authors: Himanshu Dehra

Abstract:

This paper presents a paradigm for characterization and checking of human noise behavior. The definitions of ‘Noise’ and ‘Noise Behavior’ are devised. The concept of characterization and examining of Noise Behavior is obtained from the proposed paradigm of Psychoacoustics. The measurement of human noise behavior is discussed through definitions of noise sources and noise measurements. The noise sources, noise measurement equations and noise filters are further illustrated through examples. The theory and significance of solar energy acoustics is presented for life and its activities. Human comfort and health are correlated with human brain through physiological responses and noise protection. Examples of heat stress, intense heat, sweating and evaporation are also enumerated.

Keywords: human brain, noise behavior, noise characterization, noise filters, physiological responses, psychoacoustics

Procedia PDF Downloads 507
1386 Hearing Conservation Aspects of Soldier’s Exposure to Harmfull Noise within Military Armored Vehicles

Authors: Fink Nir

Abstract:

Soldiers within armored vehicles are exposed to continuous noise reaching levels as high as 120 dB. The use of hearing protection devices (HPD) may attenuate noise by as 25 dB, but attenuated noise reaching the ear is still harmful and may result in hearing loss. Hearing conservation programs in the military suggest methods to manage the harmful effects of noise. These include noise absorption within vehicles, evaluating HPD's performance, limiting time exposure, and providing guidance.

Keywords: armored vehicle noise, hearing loss, hearing protection devices, military noise, noise attenuation

Procedia PDF Downloads 144
1385 Association of Dietary Intake with the Nutrition Knowledge, Food Label Use, and Food Preferences of Adults in San Jose del Monte City, Bulacan, Philippines

Authors: Barby Jennette A. Florano

Abstract:

Dietary intake has been associated with the health and wellbeing of adults, and lifestyle related diseases. The aim of this study was to investigate whether nutrition knowledge, food label use, and food preference are associated with the dietary intake in a sample of San Jose Del Monte City, Bulacan (SJDM) adults. A sample of 148 adults, with a mean age of 20 years, completed a validated questionnaire related to their demographic, dietary intake, nutrition knowledge, food label use and food preference. Data were analyzed using Pearson correlation and there was no association between dietary intake and nutrition knowledge. However, there were positive relationships between dietary intake and food label use (r=0.1276, p<0.10), and dietary intake and food preference (r=0.1070, p<0.10). SJDM adults who use food label and have extensive food preference had better diet quality. This finding magnifies the role of nutrition education as a potential tool in health campaigns to promote healthy eating patterns and reading food labels among students and adults. Results of this study can give information for the design of future nutrition education intervention studies to assess the efficacy of nutrition knowledge and food label use among a similar sample population.

Keywords: dietary intake, nutrition knowledge, food preference, food label use

Procedia PDF Downloads 90
1384 Exo-III Assisted Amplification Strategy through Target Recycling of Hg²⁺ Detection in Water: A GNP Based Label-Free Colorimetry Employing T-Rich Hairpin-Loop Metallobase

Authors: Abdul Ghaffar Memon, Xiao Hong Zhou, Yunpeng Xing, Ruoyu Wang, Miao He

Abstract:

Due to deleterious environmental and health effects of the Hg²⁺ ions, various online, detection methods apart from the traditional analytical tools have been developed by researchers. Biosensors especially, label, label-free, colorimetric and optical sensors have advanced with sensitive detection. However, there remains a gap of ultrasensitive quantification as noise interact significantly especially in the AuNP based label-free colorimetry. This study reported an amplification strategy using Exo-III enzyme for target recycling of Hg²⁺ ions in a T-rich hairpin loop metallobase label-free colorimetric nanosensor with an improved sensitivity using unmodified gold nanoparticles (uGNPs) as an indicator. The two T-rich metallobase hairpin loop structures as 5’- CTT TCA TAC ATA GAA AAT GTA TGT TTG -3 (HgS1), and 5’- GGC TTT GAG CGC TAA GAA A TA GCG CTC TTT G -3’ (HgS2) were tested in the study. The thermodynamic properties of HgS1 and HgS2 were calculated using online tools (http://biophysics.idtdna.com/cgi-bin/meltCalculator.cgi). The lab scale synthesized uGNPs were utilized in the analysis. The DNA sequence had T-rich bases on both tails end, which in the presence of Hg²⁺ forms a T-Hg²⁺-T mismatch, promoting the formation of dsDNA. Later, the Exo-III incubation enable the enzyme to cleave stepwise mononucleotides from the 3’ end until the structure become single-stranded. These ssDNA fragments then adsorb on the surface of AuNPs in their presence and protect AuNPs from the induced salt aggregation. The visible change in color from blue (aggregation stage in the absence of Hg²⁺) and pink (dispersion state in the presence of Hg²⁺ and adsorption of ssDNA fragments) can be observed and analyzed through UV spectrometry. An ultrasensitive quantitative nanosensor employing Exo-III assisted target recycling of mercury ions through label-free colorimetry with nanomolar detection using uGNPs have been achieved and is further under the optimization to achieve picomolar range by avoiding the influence of the environmental matrix. The proposed strategy will supplement in the direction of uGNP based ultrasensitive, rapid, onsite, label-free colorimetric detection.

Keywords: colorimetric, Exo-III, gold nanoparticles, Hg²⁺ detection, label-free, signal amplification

Procedia PDF Downloads 310
1383 Digital Forgery Detection by Signal Noise Inconsistency

Authors: Bo Liu, Chi-Man Pun

Abstract:

A novel technique for digital forgery detection by signal noise inconsistency is proposed in this paper. The forged area spliced from the other picture contains some features which may be inconsistent with the rest part of the image. Noise pattern and the level is a possible factor to reveal such inconsistency. To detect such noise discrepancies, the test picture is initially segmented into small pieces. The noise pattern and level of each segment are then estimated by using various filters. The noise features constructed in this step are utilized in energy-based graph cut to expose forged area in the final step. Experimental results show that our method provides a good illustration of regions with noise inconsistency in various scenarios.

Keywords: forgery detection, splicing forgery, noise estimation, noise

Procedia PDF Downloads 460
1382 The Improvement of Environmental Protection through Motor Vehicle Noise Abatement

Authors: Z. Jovanovic, Z. Masonicic, S. Dragutinovic, Z. Sakota

Abstract:

In this paper, a methodology for noise reduction of motor vehicles in use is presented. The methodology relies on synergic model of noise generation as a function of time. The arbitrary number of motor vehicle noise sources act in concert yielding the generation of the overall noise level of motor vehicle thereafter. The number of noise sources participating in the overall noise level of motor vehicle is subjected to the constraint of the calculation of the acoustic potential of each noise source under consideration. It is the prerequisite condition for the calculation of the acoustic potential of the whole vehicle. The recast form of pertinent set of equations describing the synergic model is laid down and solved by dint of Gauss method. The bunch of results emerged and some of them i.e. those ensuing from model application to MDD FAP Priboj motor vehicle in use are particularly elucidated.

Keywords: noise abatement, MV noise sources, noise source identification, muffler

Procedia PDF Downloads 444
1381 Evaluation of Traffic Noise Around Different Facilities Located in Silent Zones

Authors: Khaled Shaaban

Abstract:

Schools and hospitals are supposed to be located in silent zones. In these areas, it is expected to maintain low noise levels in order to promote a peaceful environment for studying or recovering. However, many of these facilities are located in urban areas and are subject to high levels of noise. In this study, an evaluation of traffic noise around schools and hospitals was conducted during different periods of the day. The results indicated that the noise is positively correlated with the traffic volume around these facilities. Locations with higher traffic volumes tend to have higher noise levels. The results also showed that the noise levels exceed the recommended values by the World Health Organization. Several solutions were suggested as potential courses of action to decrease the excessive level of noise around these facilities.

Keywords: traffic noise, road traffic, noise levels, traffic volume

Procedia PDF Downloads 99
1380 Prediction of Conducted EMI Noise in a Converter

Authors: Jon Cobb, Nasir

Abstract:

Due to higher switching frequencies, the conducted Electromagnetic interference (EMI) noise is generated in a converter. It degrades the performance of a switching converter. Therefore, it is an essential requirement to mitigate EMI noise of high performance converter. Moreover, it includes two types of emission such as common mode (CM) and differential mode (DM) noise. CM noise is due to parasitic capacitance present in a converter and DM noise is caused by switching current. However, there is dire need to understand the main cause of EMI noise. Hence, we propose a novel method to predict conducted EMI noise of different converter topologies during early stage. This paper also presents the comparison of conducted electromagnetic interference (EMI) noise due to different SMPS topologies. We also make an attempt to develop an EMI noise model for a converter which allows detailed performance analysis. The proposed method is applied to different converter, as an example, and experimental results are verified the novel prediction technique.

Keywords: EMI, electromagnetic interference, SMPS, switch-mode power supply, common mode, CM, differential mode, DM, noise

Procedia PDF Downloads 1206
1379 An Algorithm for Removal of Noise from X-Ray Images

Authors: Sajidullah Khan, Najeeb Ullah, Wang Yin Chai, Chai Soo See

Abstract:

In this paper, we propose an approach to remove impulse and Poisson noise from X-ray images. Many filters have been used for impulse noise removal from color and gray scale images with their own strengths and weaknesses but X-ray images contain Poisson noise and unfortunately there is no intelligent filter which can detect impulse and Poisson noise from X-ray images. Our proposed filter uses the upgraded layer discrimination approach to detect both Impulse and Poisson noise corrupted pixels in X-ray images and then restores only those detected pixels with a simple efficient and reliable one line equation. Our Proposed algorithms are very effective and much more efficient than all existing filters used only for Impulse noise removal. The proposed method uses a new powerful and efficient noise detection method to determine whether the pixel under observation is corrupted or noise free. Results from computer simulations are used to demonstrate pleasing performance of our proposed method.

Keywords: X-ray image de-noising, impulse noise, poisson noise, PRWF

Procedia PDF Downloads 382
1378 Development of Low Noise Savonius Wind Turbines

Authors: Sanghyeon Kim, Cheolung Cheong

Abstract:

Savonius wind turbines are a drag-type of vertical-axis wind turbine that has been used most commonly as a small-scale wind generator. However, noise is a main hindrance to wide spreading of Savonius wind turbines, just like other wind turbines. Although noise levels radiating from Savonius wind turbines may be relatively low because of their small size, they induce relatively high annoyance due to their prolonged noise exposure to the near community. Therefore, aerodynamic noise of small vertical-axis wind turbines is one of most important design parameters. In this paper, aerodynamic noise characteristics of Savonius wind turbines are investigated using the hybrid CAA techniques, and their low noise designs are proposed based on understanding of noise generation mechanism. First, flow field around the turbine are analyzed by solving 3-D unsteady incompressible RANS equations. Then, noise radiation is predicted using the Ffowcs Williams and Hawkings equation. Two distinct harmonic noise components, the well-know BPF components and the harmonics whose fundamental frequency is much higher than the BPF are identified. On a basis of this finding, S-shaped blades are proposed as low noise designs and it can reduce the noise levels of Savonius wind turbines by up to 2.7 dB.

Keywords: aerodynamic noise, Savonius wind turbine, vertical-axis wind turbine

Procedia PDF Downloads 459
1377 Mapping of Traffic Noise in Riyadh City-Saudi Arabia

Authors: Khaled A. Alsaif, Mosaad A. Foda

Abstract:

The present work aims at development of traffic noise maps for Riyadh City using the software Lima. Road traffic data were estimated or measured as accurate as possible in order to obtain consistent noise maps. The predicted noise levels at some selected sites are validated by actual field measurements, which are obtained by a system that consists of a sound level meter, a GPS receiver and a database to manage the measured data. The maps show that noise levels remain over 50 dBA and can exceed 70 dBA at the nearside of major roads and highways.

Keywords: noise pollution, road traffic noise, LimA predictor, GPS

Procedia PDF Downloads 383
1376 An Algorithm for the Map Labeling Problem with Two Kinds of Priorities

Authors: Noboru Abe, Yoshinori Amai, Toshinori Nakatake, Sumio Masuda, Kazuaki Yamaguchi

Abstract:

We consider the problem of placing labels of the points on a plane. For each point, its position, the size of its label and a priority are given. Moreover, several candidates of its label positions are prespecified, and each of such label positions is assigned a priority. The objective of our problem is to maximize the total sum of priorities of placed labels and their points. By refining a labeling algorithm that can use these priorities, we propose a new heuristic algorithm which is more suitable for treating the assigned priorities.

Keywords: map labeling, greedy algorithm, heuristic algorithm, priority

Procedia PDF Downloads 432
1375 Evaluation of Musical Conductor Exposure to Noise

Authors: Ahmed Saleh Summan

Abstract:

This article presents the results of a technical report on the evaluation of occupational noise exposures among a musical conductor in a musical rehearsal hall (party–center). A calibrated noise dosimeter was used to measure the personal exposure of a music teacher/conductor for 8 hours in two days of rehearsal involving 90 players. Results showed that noise exposure levels were much higher than the permissible levels regulated 85dBA/8hr by NIOSH. In fact, the first day of measurements recorded the highest exposure levels (91 dBA). A number of factors contributed to these results, such as players number, types of instruments used, and activities. Noise control measures were recommended to solve this situation.

Keywords: noise exposure, music conductors, occupational noise, noise in rooms

Procedia PDF Downloads 114
1374 Noise Reduction by Energising the Boundary Layer

Authors: Kiran P. Kumar, H. M. Nayana, R. Rakshitha, S. Sushmitha

Abstract:

Aircraft noise is a highly concerned problem in the field of the aviation industry. It is necessary to reduce the noise in order to be environment-friendly. Air-frame noise is caused because of the quick separation of the boundary layer over an aircraft body. So, we have to delay the boundary layer separation of an air-frame and engine nacelle. By following a certain procedure boundary layer separation can be reduced by converting laminar into turbulent and hence early separation can be prevented that leads to the noise reduction. This method has a tendency to reduce the noise of the aircraft hence it can prove efficient and environment-friendly than the present Aircraft.

Keywords: airframe, boundary layer, noise, reduction

Procedia PDF Downloads 479
1373 The Psychological Impact of Industrial Noise on Workers

Authors: Beriache Abderazik

Abstract:

It is clear that the psychological effects of noise and physiological eloquent on the workers, what will inevitably affect the performance of both productivity and efficiency in all its aspects, industrial noise became among the most prominent modern professional problems, That require study and analysis in order to arrive at solutions and ways that you can reduce the effects of industrial noise. These factors, in addition to other reasons, made us try in this research to know the real impact of industrial noise on the professional satisfaction of workers. In light of this title we have identified the following general problem: - Is the professional satisfaction factor varies depending on the noise level in the work environment? For the purpose of ascertaining the veracity of the assumptions, we have a comparative study between two samples of equal workers, the first sample is working under the influence of industrial noise severe about (100 Db), and the second sample is working under the influence of industrial noise is low (about 63 Db), and applied them test the professional satisfaction. The results support the hypotheses and confirm all sincerity.

Keywords: industrial noise, job satisfaction, the psychological effects of noise, work environment

Procedia PDF Downloads 579
1372 Scene Classification Using Hierarchy Neural Network, Directed Acyclic Graph Structure, and Label Relations

Authors: Po-Jen Chen, Jian-Jiun Ding, Hung-Wei Hsu, Chien-Yao Wang, Jia-Ching Wang

Abstract:

A more accurate scene classification algorithm using label relations and the hierarchy neural network was developed in this work. In many classification algorithms, it is assumed that the labels are mutually exclusive. This assumption is true in some specific problems, however, for scene classification, the assumption is not reasonable. Because there are a variety of objects with a photo image, it is more practical to assign multiple labels for an image. In this paper, two label relations, which are exclusive relation and hierarchical relation, were adopted in the classification process to achieve more accurate multiple label classification results. Moreover, the hierarchy neural network (hierarchy NN) is applied to classify the image and the directed acyclic graph structure is used for predicting a more reasonable result which obey exclusive and hierarchical relations. Simulations show that, with these techniques, a much more accurate scene classification result can be achieved.

Keywords: convolutional neural network, label relation, hierarchy neural network, scene classification

Procedia PDF Downloads 456
1371 Predicting Automotive Interior Noise Including Wind Noise by Statistical Energy Analysis

Authors: Yoshio Kurosawa

Abstract:

The applications of soundproof materials for reduction of high frequency automobile interior noise have been researched. This paper presents a sound pressure prediction technique including wind noise by Hybrid Statistical Energy Analysis (HSEA) in order to reduce weight of acoustic insulations. HSEA uses both analytical SEA and experimental SEA. As a result of chassis dynamo test and road test, the validity of SEA modeling was shown, and utility of the method was confirmed.

Keywords: vibration, noise, road noise, statistical energy analysis

Procedia PDF Downloads 349
1370 Traffic Noise Study at Intersection in Bangalore: A Case Study

Authors: Shiva Kumar G.

Abstract:

The present study is to know the level of noises emanated from vehicles in intersections located in urban areas using Sound Level Meter and the possibility of reducing noise levels through traffic flow optimization. The main objective is to study traffic noise level of the Intersections located at on-going metro construction activities and which are away from metro construction activities. To compare traffic noise level between stop phase, go phase and drive phase at the Intersections. To study the effect of traffic noise level of directional movement of traffic and variation in noise level during day and night times. The range of Noise level observed at intersections is between 60 to 105 decibel. The noise level of stop and drive phases were minimum and almost same where go phase had maximum noise level. By comparing noise level of directional movement of traffic, it has been noticed that Vijayanagar intersection has no significant difference in their noise level and all other intersection has a significant difference in their noise level. By comparing noise level of stop, go and drive phase it has been noticed that there was a significant difference in noise level during peak hours compared to off-peak hour. By comparing noise level between Metro and Non-Metro construction activity intersections it has been noticed that there was a significant difference in noise level. By comparing noise level during day and night times, significant differences in noise level were observed at all intersections.

Keywords: noise, metro and non-metro intersections, traffic flow optimization, stop-go and drive phase

Procedia PDF Downloads 467
1369 Modelling Railway Noise Over Large Areas, Assisted by GIS

Authors: Conrad Weber

Abstract:

The modelling of railway noise over large projects areas can be very time consuming in terms of preparing the noise models and calculation time. An open-source GIS program has been utilised to assist with the modelling of operational noise levels for 675km of railway corridor. A range of GIS algorithms were utilised to break up the noise model area into manageable calculation sizes. GIS was utilised to prepare and filter a range of noise modelling inputs, including building files, land uses and ground terrain. A spreadsheet was utilised to manage the accuracy of key input parameters, including train speeds, train types, curve corrections, bridge corrections and engine notch settings. GIS was utilised to present the final noise modelling results. This paper explains the noise modelling process and how the spreadsheet and GIS were utilised to accurately model this massive project efficiently.

Keywords: noise, modeling, GIS, rail

Procedia PDF Downloads 121
1368 Semantic Differences between Bug Labeling of Different Repositories via Machine Learning

Authors: Pooja Khanal, Huaming Zhang

Abstract:

Labeling of issues/bugs, also known as bug classification, plays a vital role in software engineering. Some known labels/classes of bugs are 'User Interface', 'Security', and 'API'. Most of the time, when a reporter reports a bug, they try to assign some predefined label to it. Those issues are reported for a project, and each project is a repository in GitHub/GitLab, which contains multiple issues. There are many software project repositories -ranging from individual projects to commercial projects. The labels assigned for different repositories may be dependent on various factors like human instinct, generalization of labels, label assignment policy followed by the reporter, etc. While the reporter of the issue may instinctively give that issue a label, another person reporting the same issue may label it differently. This way, it is not known mathematically if a label in one repository is similar or different to the label in another repository. Hence, the primary goal of this research is to find the semantic differences between bug labeling of different repositories via machine learning. Independent optimal classifiers for individual repositories are built first using the text features from the reported issues. The optimal classifiers may include a combination of multiple classifiers stacked together. Then, those classifiers are used to cross-test other repositories which leads the result to be deduced mathematically. The produce of this ongoing research includes a formalized open-source GitHub issues database that is used to deduce the similarity of the labels pertaining to the different repositories.

Keywords: bug classification, bug labels, GitHub issues, semantic differences

Procedia PDF Downloads 199
1367 Grammar as a Logic of Labeling: A Computer Model

Authors: Jacques Lamarche, Juhani Dickinson

Abstract:

This paper introduces a computational model of a Grammar as Logic of Labeling (GLL), where the lexical primitives of morphosyntax are phonological matrixes, the form of words, understood as labels that apply to realities (or targets) assumed to be outside of grammar altogether. The hypothesis is that even though a lexical label relates to its target arbitrarily, this label in a complex (constituent) label is part of a labeling pattern which, depending on its value (i.e., N, V, Adj, etc.), imposes language-specific restrictions on what it targets outside of grammar (in the world/semantics or in cognitive knowledge). Lexical forms categorized as nouns, verbs, adjectives, etc., are effectively targets of labeling patterns in use. The paper illustrates GLL through a computer model of basic patterns in English NPs. A constituent label is a binary object that encodes: i) alignment of input forms so that labels occurring at different points in time are understood as applying at once; ii) endocentric structuring - every grammatical constituent has a head label that determines the target of the constituent, and a limiter label (the non-head) that restricts this target. The N or A values are restricted to limiter label, the two differing in terms of alignment with a head. Consider the head initial DP ‘the dog’: the label ‘dog’ gets an N value because it is a limiter that is evenly aligned with the head ‘the’, restricting application of the DP. Adapting a traditional analysis of ‘the’ to GLL – apply label to something familiar – the DP targets and identifies one reality familiar to participants by applying to it the label ‘dog’ (singular). Consider next the DP ‘the large dog’: ‘large dog’ is nominal by even alignment with ‘the’, as before, and since ‘dog’ is the head of (head final) ‘large dog’, it is also nominal. The label ‘large’, however, is adjectival by narrow alignment with the head ‘dog’: it doesn’t target the head but targets a property of what dog applies to (a property or value of attribute). In other words, the internal composition of constituents determines that a form targets a property or a reality: ‘large’ and ‘dog’ happen to be valid targets to realize this constituent. In the presentation, the computer model of the analysis derives the 8 possible sequences of grammatical values with three labels after the determiner (the x y z): 1- D [ N [ N N ]]; 2- D [ A [ N N ] ]; 3- D [ N [ A N ] ]; 4- D [ A [ A N ] ]; 5- D [ [ N N ] N ]; 5- D [ [ A N ] N ]; 6- D [ [ N A ] N ] 7- [ [ N A ] N ] 8- D [ [ Adv A ] N ]. This approach that suggests that a computer model of these grammatical patterns could be used to construct ontologies/knowledge using speakers’ judgments about the validity of lexical meaning in grammatical patterns.

Keywords: syntactic theory, computational linguistics, logic and grammar, semantics, knowledge and grammar

Procedia PDF Downloads 37
1366 Effect Analysis of an Improved Adaptive Speech Noise Reduction Algorithm in Online Communication Scenarios

Authors: Xingxing Peng

Abstract:

With the development of society, there are more and more online communication scenarios such as teleconference and online education. In the process of conference communication, the quality of voice communication is a very important part, and noise may cause the communication effect of participants to be greatly reduced. Therefore, voice noise reduction has an important impact on scenarios such as voice calls. This research focuses on the key technologies of the sound transmission process. The purpose is to maintain the audio quality to the maximum so that the listener can hear clearer and smoother sound. Firstly, to solve the problem that the traditional speech enhancement algorithm is not ideal when dealing with non-stationary noise, an adaptive speech noise reduction algorithm is studied in this paper. Traditional noise estimation methods are mainly used to deal with stationary noise. In this chapter, we study the spectral characteristics of different noise types, especially the characteristics of non-stationary Burst noise, and design a noise estimator module to deal with non-stationary noise. Noise features are extracted from non-speech segments, and the noise estimation module is adjusted in real time according to different noise characteristics. This adaptive algorithm can enhance speech according to different noise characteristics, improve the performance of traditional algorithms to deal with non-stationary noise, so as to achieve better enhancement effect. The experimental results show that the algorithm proposed in this chapter is effective and can better adapt to different types of noise, so as to obtain better speech enhancement effect.

Keywords: speech noise reduction, speech enhancement, self-adaptation, Wiener filter algorithm

Procedia PDF Downloads 56
1365 Additive White Gaussian Noise Filtering from ECG by Wiener Filter and Median Filter: A Comparative Study

Authors: Hossein Javidnia, Salehe Taheri

Abstract:

The Electrocardiogram (ECG) is the recording of the heart’s electrical potential versus time. ECG signals are often contaminated with noise such as baseline wander and muscle noise. As these signals have been widely used in clinical studies to detect heart diseases, it is essential to filter these noises. In this paper we compare performance of Wiener Filtering and Median Filtering methods to filter Additive White Gaussian (AWG) noise with the determined signal to noise ratio (SNR) ranging from 3 to 5 dB applied to long-term ECG recordings samples. Root mean square error (RMSE) and coefficient of determination (R2) between the filtered ECG and original ECG was used as the filter performance indicator. Experimental results show that Wiener filter has better noise filtering performance than Median filter.

Keywords: ECG noise filtering, Wiener filtering, median filtering, Gaussian noise, filtering performance

Procedia PDF Downloads 528
1364 The Materiality of Noise Barriers: Sustainability Approach

Authors: Mostafa Gabr, Rania Abdul Galil, Nihal Salim

Abstract:

Various interventions are applied in cities with the aim to improve living and acoustic environmental conditions. Noise is one of the most influential and critical factors in the environment that has an effect on the QOL (quality of life) and urban environment. It ranks second among environmental pollution issues according to EEAA. Traffic noise is a major source of noise. Noise barriers are one of the physical techniques in landscape design used to reduce the impact of noise pollution in urban areas. Roadways noise pollution can be best controlled by a noise barrier. The aim of this paper is to consider all facets of sustainability when designing a comfortable acoustic environment in roadways, through different strategies related to planning and the design process. The study focuses on the relation between the design of noise barriers as a landscape noise mitigation installation and their materiality in so far as it influences the sustainability of the open space and the acceptability of users. According to previous studies, design of noise barrier mainly depends on cost as a decisive factor. This study asserts that environmental and socioeconomic costs associated are equally important. Hence, the paper presents a strategy for sustainable soundscape design. It builds a framework focusing on materiality considering the environmental and socioeconomic impact of noise barriers shaping urban open space around the road ways, and the different academic and market positions on noise barrier types and materials. Finally, it concludes with a matrix of the relation between the noise barrier design consideration and the three pillars of sustainability (social, economic and environmental).

Keywords: traffic noise level, acoustic sustainability, noise barrier, noise reduction, noise control, acoustical level

Procedia PDF Downloads 482