Search results for: Deep approach metacognitive methods

Authors: Firas Gerges, Frank Y. Shih

Abstract:

Malignant Melanoma, known simply as Melanoma, is a type of skin cancer that appears as a mole on the skin. It is critical to detect this cancer at an early stage because it can spread across the body and may lead to the patient death. When detected early, Melanoma is curable. In this paper we propose a deep learning model (Convolutional Neural Networks) in order to automatically classify skin lesion images as Malignant or Benign. Images underwent certain pre-processing steps to diminish the effect of the normal skin region on the model. The result of the proposed model showed a significant improvement over previous work, achieving an accuracy of 97%.

Keywords: Deep learning, skin cancer, image processing, melanoma.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1455

Authors: Sayeed Salam

Abstract:

An effort to develop a unit commitment approach capable of handling large power systems consisting of both thermal and hydro generating units offers a large profitable return. In order to be feasible, the method to be developed must be flexible, efficient and reliable. In this paper, various proposed methods have been described along with their strengths and weaknesses. As all of these methods have some sort of weaknesses, a comprehensive algorithm that combines the strengths of different methods and overcomes each other-s weaknesses would be a suitable approach for solving industry-grade unit commitment problem.

Keywords: Unit commitment, Solution methods, and Comprehensive algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 6114

Authors: M. Abdallah

Abstract:

Urban development requires deep excavations near buildings and other structures. Deep excavation has become more a necessity for better utilization of space as the population of the world has dramatically increased. In Lebanon, some urban areas are very crowded and lack spaces for new buildings and underground projects, which makes the usage of underground space indispensable. In this paper, a numerical modeling is performed using the finite element method to study the deep excavation-diaphragm wall soil-structure interaction in the case of nonlinear soil behavior. The study is focused on a comparison of the results obtained using different support systems. Furthermore, a parametric study is performed according to the remoteness of the structure.

Keywords: Deep excavation, ground anchors, interaction, struts.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1043

Authors: R. Alipour, F.Najarian

Abstract:

Because of high ductility, aluminum alloys, have been widely used as an important base of metal forming industries. But the main week point of these alloys is their low strength so in forming them with conventional methods like deep drawing, hydro forming, etc have been always faced with problems like fracture during of forming process. Because of this, recently using of explosive forming method for forming of these plates has been recommended. In this paper free explosive forming of A2024 aluminum alloy is numerically simulated and during it, explosion wave propagation process is studied. Consequences of this simulation can be effective in prediction of quality of production. These consequences are compared with an experimental test and show the superiority of this method to similar methods like hydro forming and deep drawing.

Keywords: Free explosive forming, CEL, Johnson cook.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2268

Authors: Shuen-Tai Wang, Fang-An Kuo, Chau-Yi Chou, Yu-Bin Fang

Abstract:

2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn  features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design.

Keywords: Artificial Intelligence, machine learning, deep learning, convolutional neural networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1181

Authors: Maziar Ramezani, Thomas Neitzert

Abstract:

This paper gives an overview of a deep drawing process by pressurized liquid medium separated from the sheet by a rubber diaphragm. Hydroforming deep drawing processing of sheet metal parts provides a number of advantages over conventional techniques. It generally increases the depth to diameter ratio possible in cup drawing and minimizes the thickness variation of the drawn cup. To explore the deformation mechanism, analytical and numerical simulations are used for analyzing the drawing process of an AA6061-T4 blank. The effects of key process parameters such as coefficient of friction, initial thickness of the blank and radius between cup wall and flange are investigated analytically and numerically. The simulated results were in good agreement with the results of the analytical model. According to finite element simulations, the hydroforming deep drawing method provides a more uniform thickness distribution compared to conventional deep drawing and decreases the risk of tearing during the process.

Keywords: Deep drawing, Hydroforming, Rubber diaphragm

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2859

Authors: Soofiyan Atar, Adil Shaikh, Sahil Rajpurkar, Pragnesh Bhalala, Aniket Desai, Irfan Siddavatam

Abstract:

Deep reinforcement learning (deep RL) algorithms leverage the symbolic power of complex controllers by automating it by mapping sensory inputs to low-level actions. Deep RL eliminates the complex robot dynamics with minimal engineering. Deep RL provides high-risk involvement by directly implementing it in real-world scenarios and also high sensitivity towards hyperparameters. Tuning of hyperparameters on a pneumatic quadruped robot becomes very expensive through trial-and-error learning. This paper presents an automated learning control for a pneumatic quadruped robot using sample efficient deep Q learning, enabling minimal tuning and very few trials to learn the neural network. Long training hours may degrade the pneumatic cylinder due to jerk actions originated through stochastic weights. We applied this method to the pneumatic quadruped robot, which resulted in a hopping gait. In our process, we eliminated the use of a simulator and acquired a stable gait. This approach evolves so that the resultant gait matures more sturdy towards any stochastic changes in the environment. We further show that our algorithm performed very well as compared to programmed gait using robot dynamics.

Keywords: model-based reinforcement learning, gait stability, supervised learning, pneumatic quadruped

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 526

Authors: Vishal Naranje, S. Kumar

Abstract:

In this paper a low cost knowledge base system (KBS) framework is proposed for design of deep drawing die and procedure for developing system modules. The task of building the system is structured into different modules for major activities of design of deep drawing die. A manufacturability assessment module of the proposed framework is developed to check the manufacturability of deep drawn parts. The technological knowledge is represented by using IF- THEN rules and it is coded in AutoLISP language. The module is designed to be loaded into the prompt area of AutoCAD. The cost of implementation of proposed system makes it affordable for small and medium scale sheet metal industries.

Keywords: Knowledge base system, Deep drawing die, Manufacturability, Sheet metal.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2072

Authors: K.V. Sridhar, Varun Gunnala, K.S.R Krishna Prasad

Abstract:

In this paper three basic approaches and different methods under each of them for extracting region of interest (ROI) from stationary images are explored. The results obtained for each of the proposed methods are shown, and it is demonstrated where each method outperforms the other. Two main problems in ROI extraction: the channel selection problem and the saliency reversal problem are discussed and how best these two are addressed by various methods is also seen. The basic approaches are 1) Saliency based approach 2) Wavelet based approach 3) Clustering based approach. The saliency approach performs well on images containing objects of high saturation and brightness. The wavelet based approach performs well on natural scene images that contain regions of distinct textures. The mean shift clustering approach partitions the image into regions according to the density distribution of pixel intensities. The experimental results of various methodologies show that each technique performs at different acceptable levels for various types of images.

Keywords: clustering, ROI, saliency, wavelets.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1379

Authors: Yuanqing Li, Changyi Lei, Zhaopeng Xue, Zhuo Zheng, Yanbo Long

Abstract:

In agricultural production and breeding, implementing automatic picking robot in orchard farming to reduce human labour and error is challenging. The core function of it is automatic identification based on machine vision. This paper focuses on apple detection and counting in orchards and implements several deep learning methods. Extensive datasets are used and a semi-automatic annotation method is proposed. The proposed deep learning models are in state-of-the-art YOLO family. In view of the essence of the models with various backbones, a multi-dimensional comparison in details is made in terms of counting accuracy, mAP and model memory, laying the foundation for realising automatic precision agriculture.

Keywords: Agricultural object detection, Deep learning, machine vision, YOLO family.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1008

Authors: Mannal Tariq

Abstract:

Construction industry is making progress at a high pace. The trend of the world is getting more biased towards the high rise buildings. Deep beams are one of the most common elements in modern construction having small span to depth ratio. Deep beams are mostly used as transfer girders. This experimental study consists of 16 reinforced concrete (RC) deep beams. These beams were divided into two groups; A and B. Groups A and B consist of eight beams each, having 381 mm (15 in) and 457 mm (18 in) depth respectively. Each group was further subdivided into four sub groups each consisting of two identical beams. Each subgroup was comprised of solid/control beam (without opening), opening above neutral axis (NA), at NA and below NA. Except for control beams, all beams with openings were strengthened with carbon fibre reinforced polymer (CFRP) vertical strips. These eight groups differ from each other based on depth and location of openings. For testing sake, all beams have been loaded with two symmetrical point loads. All beams have been designed based on strut and tie model concept. The outcome of experimental investigation elaborates the difference in the shear behaviour of deep beams based on depth and location of circular openings variation. 457 mm (18 in) deep beam with openings above NA show the highest strength and 381 mm (15 in) deep beam with openings below NA show the least strength. CFRP sheets played a vital role in increasing the shear capacity of beams.

Keywords: CFRP, deep beams, openings in deep beams, strut and tie model, shear behaviour.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1302

Authors: Taylan Kabbani, Ekrem Duman

Abstract:

Deep Reinforcement Learning (DRL) algorithms can scale to previously intractable problems. The automation of profit generation in the stock market is possible using DRL, by combining  the financial assets price ”prediction” step and the ”allocation” step of the portfolio in one unified process to produce fully autonomous systems capable of interacting with its environment to make optimal decisions through trial and error. This work represents a DRL model to generate profitable trades in the stock market, effectively overcoming the limitations of supervised learning approaches. We formulate the trading problem as a Partially observed Markov Decision Process (POMDP) model, considering the constraints imposed by the stock market, such as liquidity and transaction costs. We then solved the formulated POMDP problem using the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm and achieved a 2.68 Sharpe ratio on the test dataset. From the point of view of stock market forecasting and the intelligent decision-making mechanism, this paper demonstrates the superiority of DRL in financial markets over other types of machine learning and proves its credibility and advantages of strategic decision-making.

Keywords: Autonomous agent, deep reinforcement learning, MDP, sentiment analysis, stock market, technical indicators, twin delayed deep deterministic policy gradient.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 460

Authors: Muhammad Nawaz, Sandeep N. Kundu, Farha Sattar

Abstract:

Augmented reality sandbox adds new dimensions to education and learning process. It can be a core component of geoscience teaching and learning to understand the geographic contexts and landform processes. Augmented reality sandbox is a useful tool not only to create an interactive learning environment through spatial visualization but also it can provide an active learning experience to students and enhances the cognition process of learning. Augmented reality sandbox can be used as an interactive learning tool to teach geomorphic and landform processes. This article explains the augmented reality sandbox and the constructivism approach for geoscience teaching and learning, and endeavours to explore the ways to teach the geographic processes using the three-dimensional digital environment for the deep learning of the geoscience concepts interactively.

Keywords: Augmented Reality Sandbox, constructivism, deep learning, geoscience.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1474

Authors: Maxence Queyrel, Edi Prifti, Alexandre Templier, Jean-Daniel Zucker

Abstract:

Analysis of the human microbiome using metagenomic sequencing data has demonstrated high ability in discriminating various human diseases. Raw metagenomic sequencing data require multiple complex and computationally heavy bioinformatics steps prior to data analysis. Such data contain millions of short sequences read from the fragmented DNA sequences and stored as fastq files. Conventional processing pipelines consist in multiple steps including quality control, filtering, alignment of sequences against genomic catalogs (genes, species, taxonomic levels, functional pathways, etc.). These pipelines are complex to use, time consuming and rely on a large number of parameters that often provide variability and impact the estimation of the microbiome elements. Training Deep Neural Networks directly from raw sequencing data is a promising approach to bypass some of the challenges associated with mainstream bioinformatics pipelines. Most of these methods use the concept of word and sentence embeddings that create a meaningful and numerical representation of DNA sequences, while extracting features and reducing the dimensionality of the data. In this paper we present an end-to-end approach that classifies patients into disease groups directly from raw metagenomic reads: metagenome2vec. This approach is composed of four steps (i) generating a vocabulary of k-mers and learning their numerical embeddings; (ii) learning DNA sequence (read) embeddings; (iii) identifying the genome from which the sequence is most likely to come and (iv) training a multiple instance learning classifier which predicts the phenotype based on the vector representation of the raw data. An attention mechanism is applied in the network so that the model can be interpreted, assigning a weight to the influence of the prediction for each genome. Using two public real-life data-sets as well a simulated one, we demonstrated that this original approach reaches high performance, comparable with the state-of-the-art methods applied directly on processed data though mainstream bioinformatics workflows. These results are encouraging for this proof of concept work. We believe that with further dedication, the DNN models have the potential to surpass mainstream bioinformatics workflows in disease classification tasks.

Keywords: Metagenomics, phenotype prediction, deep learning, embeddings, multiple instance learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 833

Authors: S.H. Suseno, T.A.Yang, W.N. Abdullah , N.A. Febrianto, W.N. Asti, B. Bahtiar, Hamidah, A. Suman, Desniar, A. Hartoyo

Abstract:

Sixteen selected deep-sea fish obtained from Southern Java Ocean and Western Sumatra Ocean was analyzed to determine its proximate, fatty acid and mineral composition. The moisture content was ranged from 64.38 to 86.04 %, ash from 0.17 to 0.69 %, the fat content was 1.54 – 13.30 % while the protein content varied from 15.84 to 23.60%. Among the fatty acids, oleic acid and palmitic acid was the dominant MUFA and SFA. Linoleic acid was the highest PUFA found at the selected deep-sea fish. Phospor was the highest macroelement concentration on selected deep-sea fish, followed by K, Ca, Mg and Iod, Fe and Zn among microelement. The trace concentration was found at Se microelement.

Keywords: deep-sea fish, fatty acid, microelement, macroelement, monounsaturated fatty acid, proximate, polyunsaturated fatty acids.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1799

Authors: A. M. Gsiea, J. P. Goss, P. R. Briddon, Ramadan. M. Al-habashi, K. M. Etmimi, Khaled. A. S. Marghani

Abstract:

Using first-principles methods based on density functional theory and pseudopotentials, we have performed a details study of native defects in ZnO. Native point defects are unlikely to be cause of the unintentional n-type conductivity. Oxygen vacancies, which considered most often been invoked as shallow donors, have high formation energies in n-type ZnO, in edition are a deep donors. Zinc interstitials are shallow donors, with high formation energies in n-type ZnO, and thus unlikely to be responsible on their own for unintentional n-type conductivity under equilibrium conditions, as well as Zn antisites which have higher formation energies than zinc interstitials. Zinc vacancies are deep acceptors with low formation energies for n-type and in which case they will not play role in p-type coductivity of ZnO. Oxygen interstitials are stable in the form of electrically inactive split interstitials as well as deep acceptors at the octahedral interstitial site under n-type conditions. Our results may provide a guide to experimental studies of point defects in ZnO.

Keywords: DFT, Native, n-Type, ZnO.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4501

Authors: Fazıl Gökgöz, Fahrettin Filiz

Abstract:

Load forecasting has become crucial in recent years and become popular in forecasting area. Many different power forecasting models have been tried out for this purpose. Electricity load forecasting is necessary for energy policies, healthy and reliable grid systems. Effective power forecasting of renewable energy load leads the decision makers to minimize the costs of electric utilities and power plants. Forecasting tools are required that can be used to predict how much renewable energy can be utilized. The purpose of this study is to explore the effectiveness of LSTM-based neural networks for estimating renewable energy loads. In this study, we present models for predicting renewable energy loads based on deep neural networks, especially the Long Term Memory (LSTM) algorithms. Deep learning allows multiple layers of models to learn representation of data. LSTM algorithms are able to store information for long periods of time. Deep learning models have recently been used to forecast the renewable energy sources such as predicting wind and solar energy power. Historical load and weather information represent the most important variables for the inputs within the power forecasting models. The dataset contained power consumption measurements are gathered between January 2016 and December 2017 with one-hour resolution. Models use publicly available data from the Turkish Renewable Energy Resources Support Mechanism. Forecasting studies have been carried out with these data via deep neural networks approach including LSTM technique for Turkish electricity markets. 432 different models are created by changing layers cell count and dropout. The adaptive moment estimation (ADAM) algorithm is used for training as a gradient-based optimizer instead of SGD (stochastic gradient). ADAM performed better than SGD in terms of faster convergence and lower error rates. Models performance is compared according to MAE (Mean Absolute Error) and MSE (Mean Squared Error). Best five MAE results out of 432 tested models are 0.66, 0.74, 0.85 and 1.09. The forecasting performance of the proposed LSTM models gives successful results compared to literature searches.

Keywords: Deep learning, long-short-term memory, energy, renewable energy load forecasting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1536

Authors: Omer Cahana, Maya Herman, Ofer Levi

Abstract:

Magnetic Resonance Imaging (MRI) is a lengthy medical scan that stems from a long acquisition time. Its length is mainly due to the traditional sampling theorem, which defines a lower boundary for sampling. However, it is still possible to accelerate the scan by using a different approach such as Compress Sensing (CS) or Parallel Imaging (PI). These two complementary methods can be combined to achieve a faster scan with high-fidelity imaging. To achieve that, two conditions must be satisfied: i) the signal must be sparse under a known transform domain, and ii) the sampling method must be incoherent. In addition, a nonlinear reconstruction algorithm must be applied to recover the signal. While the rapid advances in Deep Learning (DL) have had tremendous successes in various computer vision tasks, the field of MRI reconstruction is still in its early stages. In this paper, we present an end-to-end method for MRI reconstruction from k-space to image. Our method contains two parts. The first is sensitivity map estimation (SME), which is a small yet effective network that can easily be extended to a variable number of coils. The second is reconstruction, which is a top-down architecture with lateral connections developed for building high-level refinement at all scales. Our method holds the state-of-art fastMRI benchmark, which is the largest, most diverse benchmark for MRI reconstruction.

Keywords: Accelerate MRI scans, image reconstruction, pyramid network, deep learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 263

Authors: T. Damak, O. Kriaa, A. Baccar, M. A. Ben Ayed, N. Masmoudi

Abstract:

In the last few years, Automatic Number Plate Recognition (ANPR) systems have become widely used in the safety, the security, and the commercial aspects. Forethought, several methods and techniques are computing to achieve the better levels in terms of accuracy and real time execution. This paper proposed a computer vision algorithm of Number Plate Localization (NPL) and Characters Segmentation (CS). In addition, it proposed an improved method in Optical Character Recognition (OCR) based on Deep Learning (DL) techniques. In order to identify the number of detected plate after NPL and CS steps, the Convolutional Neural Network (CNN) algorithm is proposed. A DL model is developed using four convolution layers, two layers of Maxpooling, and six layers of fully connected. The model was trained by number image database on the Jetson TX2 NVIDIA target. The accuracy result has achieved 95.84%.

Keywords: Automatic number plate recognition, character segmentation, convolutional neural network, CNN, deep learning, number plate localization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1215

Authors: Zhongjie Wang, Xiaofei Xu, Dechen Zhan

Abstract:

With deep development of software reuse, componentrelated technologies have been widely applied in the development of large-scale complex applications. Component identification (CI) is one of the primary research problems in software reuse, by analyzing domain business models to get a set of business components with high reuse value and good reuse performance to support effective reuse. Based on the concept and classification of CI, its technical stack is briefly discussed from four views, i.e., form of input business models, identification goals, identification strategies, and identification process. Then various CI methods presented in literatures are classified into four types, i.e., domain analysis based methods, cohesion-coupling based clustering methods, CRUD matrix based methods, and other methods, with the comparisons between these methods for their advantages and disadvantages. Additionally, some insufficiencies of study on CI are discussed, and the causes are explained subsequently. Finally, it is concluded with some significantly promising tendency about research on this problem.

Keywords: Business component, component granularity, component identification, reuse performance.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1928

Authors: Joan Antony

Abstract:

Outcome in any learning process should target three goals – propelling the underachiever’s engagement in the learning process, enhancing the drive to achieve, and modifying attitudes and beliefs in his/her capabilities. An intervention study with a three-pronged approach incorporating self-regulatory training targeting three categories of strategies – cognitive, metacognitive and motivational – was designed adopting the before and after control-experimental group design. The evaluation of the training process was based on pre- and post-intervention measures obtained through three indices of measurement – academic scores based on grades on school examinations and comprehension tests, affective variables scores and level of strategy use obtained through responses on scales and questionnaires, and content analysis of subjective responses to open-ended probes. The evaluation relied on three sources – student, teacher and parent. The t-test results for the experimental and control groups on the pre- and post-intervention measurements indicate a significant increase on comprehension tasks for the experimental group. Though statistically significant difference was not found on the school examination scores for the experimental group, there was considerable decline in performance for the control group. Analysis of covariance (ANCOVA) was applied on the scores obtained on affective variables, namely, self-esteem, personal achievement goals, personal ego goals, personal task goals, and locus of control. The experimental group showed increase in personal achievement goals and personal ego goals as compared to the control group. Responses given by the experimental group to the open-ended probes on causal attributions indicated a considerable shift from external to internal causes when moving from the pre- to post-intervention stage. ANCOVA results revealed significantly higher use of learning strategies inclusive of mental learning strategies, behavioral learning strategies, self-regulatory strategies, and an improvement in study orientation encompassing study habits and study attitudes among the experimental group students. Parents and teachers reported significant progressive transformation towards constructive engagement with study material and self-imposed regulation. The implications of this study are three-fold: firstly, strategies training (cognitive, metacognitive and motivational) should be embedded into daily classroom routine; secondly, scaffolding by teachers through activities based on curriculum will eventually enable students to rely more on their own judgements of effective strategy use; thirdly, enhanced confidence will radiate to the affective aspects with enduring effects on other domains of life as well. The cyclic nature of the interaction between utilizing one’s resources, managing effort and regulating emotions forms the foundation for academic achievement.

Keywords: Academic achievement, cognitive strategies, metacognitive strategies, motivational strategies.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 406

Authors: Jie Zhang, Qianyu Guo, Tieyi Zhang, Zhiyong Feng, Xiaohong Li

Abstract:

The widespread popularity of mobile devices and the development of artificial intelligence (AI) have led to the widespread adoption of deep learning (DL) in Android apps. Compared with traditional Android apps (traditional apps), deep learning based Android apps (DL-based apps) need to use more third-party application programming interfaces (APIs) to complete complex DL inference tasks. However, existing methods (e.g., FlowDroid) for detecting sensitive information leakage in Android apps cannot be directly used to detect DL-based apps as they are difficult to detect third-party APIs. To solve this problem, we design DLtrace, a new static information flow analysis tool that can effectively recognize third-party APIs. With our proposed trace and detection algorithms, DLtrace can also efficiently detect privacy leaks caused by sensitive APIs in DL-based apps. Additionally, we propose two formal definitions to deal with the common polymorphism and anonymous inner-class problems in the Android static analyzer. Using DLtrace, we summarize the non-sequential characteristics of DL inference tasks in DL-based apps and the specific functionalities provided by DL models for such apps. We conduct an empirical assessment with DLtrace on 208 popular DL-based apps in the wild and found that 26.0% of the apps suffered from sensitive information leakage. Furthermore, DLtrace outperformed FlowDroid in detecting and identifying third-party APIs. The experimental results demonstrate that DLtrace expands FlowDroid in understanding DL-based apps and detecting security issues therein.

Keywords: Mobile computing, deep learning apps, sensitive information, static analysis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 496

Authors: Delowar Hossain, Genci Capi

Abstract:

This paper concerns with the problem of deep learning parameters tuning using a genetic algorithm (GA) in order to improve the performance of deep learning (DL) method. We present a GA based DL method for robot object recognition and grasping. GA is used to optimize the DL parameters in learning procedure in term of the fitness function that is good enough. After finishing the evolution process, we receive the optimal number of DL parameters. To evaluate the performance of our method, we consider the object recognition and robot grasping tasks. Experimental results show that our method is efficient for robot object recognition and grasping.

Keywords: Deep learning, genetic algorithm, object recognition, robot grasping.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2084

Authors: Ming-Jen Huang, Chun-Fang Huang, Chiching Wei

Abstract:

With the recent advance of the deep neural network, we observe new applications of NLP (natural language processing) and CV (computer vision) powered by deep neural networks for processing business documents. However, creating a real-world document processing system needs to integrate several NLP and CV tasks, rather than treating them separately. There is a need to have a unified approach for processing documents containing textual and graphical elements with rich formats, diverse layout arrangement, and distinct semantics. In this paper, a framework that fulfills this unified approach is presented. The framework includes a representation model definition for holding the information generated by various tasks and specifications defining the coordination between these tasks. The framework is a blueprint for building a system that can process documents with rich formats, styles, and multiple types of elements. The flexible and lightweight design of the framework can help build a system for diverse business scenarios, such as contract monitoring and reviewing.

Keywords: Document processing, framework, formal definition, machine learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 579

Authors: D. I. Ekine, C. C. Akpanibah

Abstract:

This paper analyses the productivity of median scale battery cage and deep litter chicken egg producers in Rivers State, Nigeria. 90 battery cage and 90 deep litter farmers giving a total of 180 farmers were sampled through a multistage sampling procedure. Mean productivity was higher for the battery cage than the deep litter farmers at 2.65 and 2.33 respectively. Productivity of battery cage farmers were positively influenced by age, extension contacts, experience and feed quantity while the productivity of deep litter farmers was positively influenced by age, extension contacts, household size, experience and labour. The major constraints identified by both categories are high cost of feed, high price of day-old chick, inadequate finance, lack of credit and high cost of drug/vaccination. Furthermore, the work recommends that government should assist chicken egg farmers through subsidies of input resources and put policies to make financial institutions give out loans at low interest rate to the farmers. The farmers should abide by the recommended number of birds per unit area while stocking.

Keywords: Productivity, battery cage, deep litter, median scale, egg production.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 178

Authors: Jia Xin Low, Keng Wah Choo

Abstract:

This paper presents an automatic normal and abnormal heart sound classification model developed based on deep learning algorithm. MITHSDB heart sounds datasets obtained from the 2016 PhysioNet/Computing in Cardiology Challenge database were used in this research with the assumption that the electrocardiograms (ECG) were recorded simultaneously with the heart sounds (phonocardiogram, PCG). The PCG time series are segmented per heart beat, and each sub-segment is converted to form a square intensity matrix, and classified using convolutional neural network (CNN) models. This approach removes the need to provide classification features for the supervised machine learning algorithm. Instead, the features are determined automatically through training, from the time series provided. The result proves that the prediction model is able to provide reasonable and comparable classification accuracy despite simple implementation. This approach can be used for real-time classification of heart sounds in Internet of Medical Things (IoMT), e.g. remote monitoring applications of PCG signal.

Keywords: Convolutional neural network, discrete wavelet transform, deep learning, heart sound classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1103

Authors: M. V. A. Raju Bahubalendruni, Bibhuti Bhusan Biswal, B. B. V. L. Deepak

Abstract:

Achieving an appropriate assembly sequence needs deep verification for its physical feasibility. For this purpose, industrial engineers use several assembly predicates; namely, liaison, geometric feasibility, stability and mechanical feasibility. However, testing an assembly sequence for these predicates requires huge assembly information. Extracting such assembly information from an assembled product is a time consuming and highly skillful task with complex reasoning methods. In this paper, computer aided methods are proposed to extract all the necessary assembly information from computer aided design (CAD) environment in order to perform the assembly sequence planning efficiently. These methods use preliminary capabilities of three-dimensional solid modelling and assembly modelling methods used in CAD software considering equilibrium laws of physical bodies.

Keywords: Assembly automation, assembly attributes, assembly sequence generation, computer aided design.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1268

Authors: H.Seidi, M.Jalali Azizpour, S.A.Zahedi

Abstract:

Today, Hydroforming technology provides an attractive alternative to conventional matched die forming, especially for cost-sensitive, lower volume production, and for parts with irregular contours. In this study the critical fluid pressures which lead to rupture in the workpiece has been investigated by theoretical and finite element methods. The axisymmetric analysis was developed to investigate the tearing phenomenon in cylindrical Hydroforming Deep Drawing (HDD). By use of obtained equations the effect of anisotropy, drawing ratio, sheet thickness and strain hardening exponent on tearing diagram were investigated.

Keywords: Hydroforming deep drawing, Pressure path, Axisymmetric analysis, Finite element simulation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1626

Authors: M. A. Sanda, B. Johansson, J. Johansson

Abstract:

The purpose of this paper is to conceptualize a futureoriented human work environment and organizational activity in deep mines that entails a vision of good and safe workplace. Futureoriented technological challenges and mental images required for modern work organization design were appraised. It is argued that an intelligent-deep-mine covering the entire value chain, including environmental issues and with work organization that supports good working and social conditions towards increased human productivity could be designed. With such intelligent system and work organization in place, the mining industry could be seen as a place where cooperation, skills development and gender equality are key components. By this perspective, both the youth and women might view mining activity as an attractive job and the work environment as a safe, and this could go a long way in breaking the unequal gender balance that exists in most mines today.

Keywords: Mining activity; deep mining; human operators; intelligent deep mine; work environment; organizational activity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1610

Authors: Ahmed, M., Yasser A., Mahmoud H., Ahmed, A., Abdulla M. S., Nazar, S.

Abstract:

Strut-and-Tie Models (STM) for the design of concrete beams, comprising of struts, ties, nodes as the basic tools, is conceptually simple, but its realization for complex concrete structure is not straightforward and depends on flow of internal forces in the structure. STM technique has won wide acceptance for deep member and shear design. STM technique is a unified approach that considers all load effects (bending, axial, shear, and torsion) simultaneously, not just applicable to shear loading only. The present study is to portray Strut-and-Tie Modeling based on Load-Transfer-Mechanisms as a unified method to analyze, design and detailing for deep and slender concrete beams. Three shear span- effective depth ratio (a/ d) are recommended for the modeling of STM elements corresponding to dominant load paths. The study also discusses the research work conduct on effective stress of concrete, tie end anchorage, and transverse reinforcement demand under different load transfer mechanism. It is also highlighted that to make the STM versatile tool for design of beams applicable to all shear spans, the effective stress of concrete and, transverse reinforcement demand, inclined angle of strut, and anchorage requirements of tie bars is required to be correlated with respect to load transfer mechanism. The country code provisions are to be modified and updated to apply for generalized design of concrete deep and slender member using load transfer mechanism based STM technique. Examples available in literature are reanalyzed with refined STM based on load transfer mechanisms and results are compared. It is concluded from the results that proposed approach will require true reinforcement demand depending on dominant force transfer action in concrete beam.

Keywords: Deep member, Load transfer mechanism, Strut-and-Tie Model, Strut, Truss.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5939