Search results for: algorithm optimisation

Authors: M. S. Wickramarachchi, L. S. Nawarathna, C. M. B. Dematawewa

Abstract:

Hatchery performance is critical for the profitability of poultry breeder operations. Some extrinsic parameters of eggs and breeders cause to increase or decrease the hatchability. This study aims to identify the affecting extrinsic parameters on the commercial hatchability of local chicken's eggs and determine the most efficient classification model with a hatchability rate greater than 90%. In this study, seven extrinsic parameters were considered: egg weight, moisture loss, breeders age, number of fertilised eggs, shell width, shell length, and shell thickness. Multiple linear regression was performed to determine the most influencing variable on hatchability. First, the correlation between each parameter and hatchability were checked. Then a multiple regression model was developed, and the accuracy of the fitted model was evaluated. Linear Discriminant Analysis (LDA), Classification and Regression Trees (CART), k-Nearest Neighbors (kNN), Support Vector Machines (SVM) with a linear kernel, and Random Forest (RF) algorithms were applied to classify the hatchability. This grouping process was conducted using binary classification techniques. Hatchability was negatively correlated with egg weight, breeders' age, shell width, shell length, and positive correlations were identified with moisture loss, number of fertilised eggs, and shell thickness. Multiple linear regression models were more accurate than single linear models regarding the highest coefficient of determination (R²) with 94% and minimum AIC and BIC values. According to the classification results, RF, CART, and kNN had performed the highest accuracy values 0.99, 0.975, and 0.972, respectively, for the commercial hatchery process. Therefore, the RF is the most appropriate machine learning algorithm for classifying the breeder outcomes, which are economically profitable or not, in a commercial hatchery.

Keywords: classification models, egg weight, fertilised eggs, multiple linear regression

Procedia PDF Downloads 87

Authors: Kaiwen Zheng, Wanting Zhou, Nan Tang, Lei Li, Yuanhang He

Abstract:

The integrated circuit industry has become a cornerstone of the information society, finding widespread application in areas such as industry, communication, medicine, and aerospace. However, with the increasing complexity of integrated circuits, Hardware Trojans (HTs) implanted by attackers have become a significant threat to their security. In this paper, we proposed a hardware trojan detection method for large-scale circuits. As HTs introduce physical characteristic changes such as structure, area, and power consumption as additional redundant circuits, we proposed a machine-learning-based hardware trojan detection method based on the physical characteristics of gate-level netlists. This method transforms the hardware trojan detection problem into a machine-learning binary classification problem based on physical characteristics, greatly improving detection speed. To address the problem of imbalanced data, where the number of pure circuit samples is far less than that of HTs circuit samples, we used the SMOTETomek algorithm to expand the dataset and further improve the performance of the classifier. We used three machine learning algorithms, K-Nearest Neighbors, Random Forest, and Support Vector Machine, to train and validate benchmark circuits on Trust-Hub, and all achieved good results. In our case studies based on AES encryption circuits provided by trust-hub, the test results showed the effectiveness of the proposed method. To further validate the method’s effectiveness for detecting variant HTs, we designed variant HTs using open-source HTs. The proposed method can guarantee robust detection accuracy in the millisecond level detection time for IC, and FPGA design flows and has good detection performance for library variant HTs.

Keywords: hardware trojans, physical properties, machine learning, hardware security

Procedia PDF Downloads 146

Authors: Oumaima Khlifati, Khadija Baba

Abstract:

Pavement distress is the main factor responsible for the deterioration of road structure durability, damage vehicles, and driver comfort. Transportation agencies spend a high proportion of their funds on pavement monitoring and maintenance. The auscultation of pavement distress was based on the manual survey, which was extremely time consuming, labor intensive, and required domain expertise. Therefore, the automatic distress detection is needed to reduce the cost of manual inspection and avoid more serious damage by implementing the appropriate remediation actions at the right time. Inspired by recent deep learning applications, this paper proposes an algorithm for automatic road distress detection and classification using on the Deep Convolutional Neural Network (DCNN). In this study, the types of pavement distress are classified as transverse or longitudinal cracking, alligator, pothole, and intact pavement. The dataset used in this work is composed of public asphalt pavement images. In order to learn the structure of the different type of distress, the DCNN models are trained and tested as a multi-label classification task. In addition, to get the highest accuracy for our model, we adjust the structural optimization hyper parameters such as the number of convolutions and max pooling, filers, size of filters, loss functions, activation functions, and optimizer and fine-tuning hyper parameters that conclude batch size and learning rate. The optimization of the model is executed by checking all feasible combinations and selecting the best performing one. The model, after being optimized, performance metrics is calculated, which describe the training and validation accuracies, precision, recall, and F1 score.

Keywords: distress pavement, hyperparameters, automatic classification, deep learning

Procedia PDF Downloads 93

Authors: Abdulmohsen Alruwaili

Abstract:

A partial differential system featuring momentum and energy balance is often used to describe simulations of flow initiation and thermal shifting in boundary layers. The buoyancy force in terms of temperature is factored in the momentum balance equation. Buoyancy force causes the flow quantity to fluctuate along the streamwise direction 𝑋; therefore, the problem can be, to our best knowledge, analyzed through nonsimilar modeling. In this analysis, a nonsimilar model is evolved for radiative mixed convection of a magnetized power-law nanoliquid flow on top of a vertical plate installed in a stationary fluid. The upward linear stretching initiated the flow in the vertical direction. Assuming nanofluids are composite of copper (Cu) and alumina (Al₂O₃) nanoparticles, the viscous dissipation in this case is negligible. The nonsimilar system is dealt with analytically by local nonsimilarity (LNS) via numerical algorithm bvp4c. Surface temperature and flow field are shown visually in relation to factors like mixed convection, magnetic field strength, nanoparticle volume fraction, radiation parameters, and Prandtl number. The repercussions of magnetic and mixed convection parameters on the rate of energy transfer and friction coefficient are represented in tabular forms. The results obtained are compared to the published literature. It is found that the existence of nanoparticles significantly improves the temperature profile of considered nanoliquid. It is also observed that when the estimates of the magnetic parameter increase, the velocity profile decreases. Enhancement in nanoparticle concentration and mixed convection parameter improves the velocity profile.

Keywords: nanofluid, power law model, mixed convection, thermal radiation

Procedia PDF Downloads 32

Authors: Leizhang Wang, Baocang Wang

Abstract:

General Sieve Kernel (G6K) is considered as currently the fastest algorithm for the shortest vector problem (SVP) and record holder of open SVP challenge. We study the lattice basis quality improvement effects of the Workout proposed in G6K, which is composed of a series of pumps to solve SVP. Firstly, we use a low-dimensional pump output basis to propose a predictor to predict the quality of high-dimensional Pumps output basis. Both theoretical analysis and experimental tests are performed to illustrate that it is more computationally expensive to solve the LWE problems by using a G6K default SVP solving strategy (Workout) than these lattice reduction algorithms (e.g. BKZ 2.0, Progressive BKZ, Pump, and Jump BKZ) with sieving as their SVP oracle. Secondly, the default Workout in G6K is optimized to achieve a stronger reduction and lower computational cost. Thirdly, we combine the optimized Workout and the Pump output basis quality predictor to further reduce the computational cost by optimizing LWE instances selection strategy. In fact, we can solve the TU LWE challenge (n = 65, q = 4225, = 0:005) 13.6 times faster than the G6K default Workout. Fourthly, we consider a combined two-stage (Preprocessing by BKZ- and a big Pump) LWE solving strategy. Both stages use dimension for free technology to give new theoretical security estimations of several LWE-based cryptographic schemes. The security estimations show that the securities of these schemes with the conservative Newhope’s core-SVP model are somewhat overestimated. In addition, in the case of LAC scheme, LWE instances selection strategy can be optimized to further improve the LWE-solving efficiency even by 15% and 57%. Finally, some experiments are implemented to examine the effects of our strategies on the Normal Form LWE problems, and the results demonstrate that the combined strategy is four times faster than that of Newhope.

Keywords: LWE, G6K, pump estimator, LWE instances selection strategy, dimension for free

Procedia PDF Downloads 60

Authors: Biruk Wolde, Atalay Ayele, Yonatan Garkabo, Trufat Hailmariam, Zemenu Germewu

Abstract:

A comprehensive three-dimensional (3D) geological modeling and engineering geological characterization of shallow subsurface soils and rocks are essential for a wide range of geotechnical and seismological engineering applications, particularly in urban environments. The spatial distribution and geological variation of the shallow subsurface of Addis Ababa city have not been studied so far in terms of geological and geotechnical modeling. This study aims at the construction of a 3D geological model, as well as provides awareness into the engineering geological characteristics of shallow subsurface soil and rock of Addis Ababa city. The 3D geological model was constructed by using more than 1500 geotechnical boreholes, well-drilling data, and geological maps. A well-known geostatistical kriging 3D interpolation algorithm was applied to visualize the spatial distribution and geological variation of the shallow subsurface. Due to the complex nature of geological formations, vertical and lateral variation of the geological profiles horizons-solid command has been selected via the Groundwater Modelling System (GMS) graphical user interface software. For the engineering geological characterization of typical soils and rocks, both index and engineering laboratory tests have been used. The geotechnical properties of soil and rocks vary from place to place due to the uneven nature of subsurface formations observed in the study areas. The constructed model ascertains the thickness, extent, and 3D distribution of the important geological units of the city. This study is the first comprehensive research work on 3D geological modeling and subsurface characterization of soils and rocks in Addis Ababa city, and the outcomes will be important for further future research on subsurface conditions in the city. Furthermore, these findings provide a reference for developing a geo-database for the city.

Keywords: 3d geological modeling, addis ababa, engineering geology, geostatistics, horizons-solid

Procedia PDF Downloads 98

Authors: Chun-Min Lin, Yuh-Jen Wu, Ching-Ting Chung

Abstract:

The formulation of tourism policies requires objective academic research and evidence as support, especially research from local academia. Taiwan is a small island, and its economic growth relies heavily on tourism revenue. Taiwanese government has been devoting to the promotion of the tourism industry over the past few decades. Scientific research outcomes by Taiwanese scholars may and will help lay the foundations for drafting future tourism policy by the government. In this study, a total of 120 full journal articles published between 2008 and 2016 from the Journal of Tourism and Leisure Studies (JTSL) were examined to explore the scientific research trend of tourism study in Taiwan. JTSL is one of the most important Taiwanese journals in the tourism discipline which focuses on tourism-related issues and uses traditional Chinese as the study language. The method of co-word analysis from bibliometrics approaches was employed for semantic analysis in this study. When analyzing Chinese words and phrases, word segmentation analysis is a crucial step. It must be carried out initially and precisely in order to obtain meaningful word or word chunks for further frequency calculation. A word segmentation system basing on N-gram algorithm was developed in this study to conduct semantic analysis, and 100 groups of meaningful phrases with the highest recurrent rates were located. Subsequently, co-word analysis was employed for semantic classification. The results showed that the themes of tourism research in Taiwan in recent years cover the scope of tourism education, environmental protection, hotel management, information technology, and senior tourism. The results can give insight on the related issues and serve as a reference for tourism-related policy making and follow-up research.

Keywords: bibliometrics, co-word analysis, word segmentation, tourism research, policy

Procedia PDF Downloads 229

Authors: Y. Trantafyllou, C. Alexandraki

Abstract:

Music interpretation accounts to the way musicians shape their performance by deliberately deviating from composers’ intentions, which are commonly communicated via some form of music transcription, such as a music score. For transcribed and non-improvised music, music expression is manifested by introducing subtle deviations in tempo, dynamics and articulation during the evolution of performance. This paper presents an application, named FlameCens, which, given two recordings of the same piece of music, presumably performed by different musicians, allow visualising deviations in tempo and dynamics during playback. The application may also compare a certain performance to the music score of that piece (i.e. MIDI file), which may be thought of as an expression-neutral representation of that piece, hence depicting the expressive queues employed by certain performers. FlameCens uses the Dynamic Time Warping algorithm to compare two audio sequences, based on CENS (Chroma Energy distribution Normalized Statistics) audio features. Expressive deviations are illustrated in a moving flame, which is generated by an animation of particles. The length of the flame is mapped to deviations in dynamics, while the slope of the flame is mapped to tempo deviations so that faster tempo changes the slope to the right and slower tempo changes the slope to the left. Constant slope signifies no tempo deviation. The detected deviations in tempo and dynamics can be additionally recorded in a text file, which allows for offline investigation. Moreover, in the case of monophonic music, the color of particles is used to convey the pitch of the notes during performance. FlameCens has been implemented in Python and it is openly available via GitHub. The application has been experimentally validated for different music genres including classical, contemporary, jazz and popular music. These experiments revealed that FlameCens can be a valuable tool for music specialists (i.e. musicians or musicologists) to investigate the expressive performance strategies employed by different musicians, as well as for music audience to enhance their listening experience.

Keywords: audio synchronization, computational music analysis, expressive music performance, information visualization

Procedia PDF Downloads 130

Authors: V. Veeraprathap, G. S. Harish, G. Narendra Kumar

Abstract:

Uncontrolled growth of abnormal cells in the lung in the form of tumor can be either benign (non-cancerous) or malignant (cancerous). Patients with Lung Cancer (LC) have an average of five years life span expectancy provided diagnosis, detection and prediction, which reduces many treatment options to risk of invasive surgery increasing survival rate. Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI) for earlier detection of cancer are common. Gaussian filter along with median filter used for smoothing and noise removal, Histogram Equalization (HE) for image enhancement gives the best results without inviting further opinions. Lung cavities are extracted and the background portion other than two lung cavities is completely removed with right and left lungs segmented separately. Region properties measurements area, perimeter, diameter, centroid and eccentricity measured for the tumor segmented image, while texture is characterized by Gray-Level Co-occurrence Matrix (GLCM) functions, feature extraction provides Region of Interest (ROI) given as input to classifier. Two levels of classifications, K-Nearest Neighbor (KNN) is used for determining patient condition as normal or abnormal, while Artificial Neural Networks (ANN) is used for identifying the cancer stage is employed. Discrete Wavelet Transform (DWT) algorithm is used for the main feature extraction leading to best efficiency. The developed technology finds encouraging results for real time information and on line detection for future research.

Keywords: artificial neural networks, ANN, discrete wavelet transform, DWT, gray-level co-occurrence matrix, GLCM, k-nearest neighbor, KNN, region of interest, ROI

Procedia PDF Downloads 153

Authors: Apostolos Kalatzis, Ashish Teotia, Vishnunarayan Girishan Prabhu, Laura Stanley

Abstract:

Emotions are functional behaviors influenced by thoughts, stimuli, and other factors that induce neurophysiological changes in the human body. Understanding and classifying emotions are challenging as individuals have varying perceptions of their environments. Therefore, it is crucial that there are publicly available databases and virtual reality (VR) based environments that have been scientifically validated for assessing emotional classification. This study utilized two commercially available VR applications (Guided Meditation VR™ and Richie’s Plank Experience™) to induce acute stress and calm state among participants. Subjective and objective measures were collected to create a validated multimodal dataset and classification scheme for affective state classification. Participants’ subjective measures included the use of the Self-Assessment Manikin, emotional cards and 9 point Visual Analogue Scale for perceived stress, collected using a Virtual Reality Assessment Tool developed by our team. Participants’ objective measures included Electrocardiogram and Respiration data that were collected from 25 participants (15 M, 10 F, Mean = 22.28  4.92). The features extracted from these data included heart rate variability components and respiration rate, both of which were used to train two machine learning models. Subjective responses validated the efficacy of the VR applications in eliciting the two desired affective states; for classifying the affective states, a logistic regression (LR) and a support vector machine (SVM) with a linear kernel algorithm were developed. The LR outperformed the SVM and achieved 93.8%, 96.2%, 93.8% leave one subject out cross-validation accuracy, precision and recall, respectively. The VR assessment tool and data collected in this study are publicly available for other researchers.

Keywords: affective computing, biosignals, machine learning, stress database

Procedia PDF Downloads 142

Authors: M. Bilal Ishfaq, Adnan N. Qureshi

Abstract:

COVID-19 is the most infectious disease these days, it was first reported in Wuhan, the capital city of Hubei in China then it spread rapidly throughout the whole world. Later on 11 March 2020, the World Health Organisation (WHO) declared it a pandemic. Since COVID-19 is highly contagious, it has affected approximately 219M people worldwide and caused 4.55M deaths. It has brought the importance of accurate diagnosis of respiratory diseases such as pneumonia and COVID-19 to the forefront. In this paper, we propose a hybrid approach for the automated detection of COVID-19 using medical imaging. We have presented the hybridization of manually extracted and convolutional features. Our approach combines Haralick texture features and convolutional features extracted from chest X-rays and CT scans. We also employ a minimum redundancy maximum relevance (MRMR) feature selection algorithm to reduce computational complexity and enhance classification performance. The proposed model is evaluated on four publicly available datasets, including Chest X-ray Pneumonia, COVID-19 Pneumonia, COVID-19 CTMaster, and VinBig data. The results demonstrate high accuracy and effectiveness, with 0.9925 on the Chest X-ray pneumonia dataset, 0.9895 on the COVID-19, Pneumonia and Normal Chest X-ray dataset, 0.9806 on the Covid CTMaster dataset, and 0.9398 on the VinBig dataset. We further evaluate the effectiveness of the proposed model using ROC curves, where the AUC for the best-performing model reaches 0.96. Our proposed model provides a promising tool for the early detection and accurate diagnosis of COVID-19, which can assist healthcare professionals in making informed treatment decisions and improving patient outcomes. The results of the proposed model are quite plausible and the system can be deployed in a clinical or research setting to assist in the diagnosis of COVID-19.

Keywords: COVID-19, feature engineering, artificial neural networks, radiology images

Procedia PDF Downloads 75

Authors: Meng H. Lean, Wei-Ping L. Chu

Abstract:

The effects of ferroelectric nanofiller size, shape, loading, and polarization, on bipolar charge injection, transport, and recombination through amorphous and semicrystalline polymers are studied. A 3D particle-in-cell model extends the classical electrical double layer representation to treat ferroelectric nanoparticles. Metal-polymer charge injection assumes Schottky emission and Fowler-Nordheim tunneling, migration through field-dependent Poole-Frenkel mobility, and recombination with Monte Carlo selection based on collision probability. A boundary integral equation method is used for solution of the Poisson equation coupled with a second-order predictor-corrector scheme for robust time integration of the equations of motion. The stability criterion of the explicit algorithm conforms to the Courant-Friedrichs-Levy limit. Trajectories for charge that make it through the film are curvilinear paths that meander through the interspaces. Results indicate that charge transport behavior depends on nanoparticle polarization with anti-parallel orientation showing the highest leakage conduction and lowest level of charge trapping in the interaction zone. Simulation prediction of a size range of 80 to 100 nm to minimize attachment and maximize conduction is validated by theory. Attached charge fractions go from 2.2% to 97% as nanofiller size is decreased from 150 nm to 60 nm. Computed conductivity of 0.4 x 1014 S/cm is in agreement with published data for plastics. Charge attachment is increased with spheroids due to the increase in surface area, and especially so for oblate spheroids showing the influence of larger cross-sections. Charge attachment to nanofillers and nanocrystallites increase with vol.% loading or degree of crystallinity, and saturate at about 40 vol.%.

Keywords: nanocomposites, nanofillers, electrical double layer, bipolar charge transport

Procedia PDF Downloads 354

Authors: Darwin Castillo Huamaní, Francisco A. M. Gomes

Abstract:

The aim of the multi-material topology optimization (MTO) is to obtain the optimal topology of structures composed by many materials, according to a given set of constraints and cost criteria. In this work, we seek the optimal distribution of materials in a domain, such that the flexibility of the structure is minimized, under certain boundary conditions and the intervention of external forces. In the case we have only one material, each point of the discretized domain is represented by two values from a function, where the value of the function is 1 if the element belongs to the structure or 0 if the element is empty. A common way to avoid the high computational cost of solving integer variable optimization problems is to adopt the Solid Isotropic Material with Penalization (SIMP) method. This method relies on the continuous interpolation function, power function, where the base variable represents a pseudo density at each point of domain. For proper exponent values, the SIMP method reduces intermediate densities, since values other than 0 or 1 usually does not have a physical meaning for the problem. Several extension of the SIMP method were proposed for the multi-material case. The one that we explore here is the ordered SIMP method, that has the advantage of not being based on the addition of variables to represent material selection, so the computational cost is independent of the number of materials considered. Although the number of variables is not increased by this algorithm, the optimization subproblems that are generated at each iteration cannot be solved by methods that rely on second derivatives, due to the cost of calculating the second derivatives. To overcome this, we apply a globally convergent version of the sequential linear programming method, which solves a linear approximation sequence of optimization problems.

Keywords: globally convergence, multi-material design ordered simp, sequential linear programming, topology optimization

Procedia PDF Downloads 315

Authors: Abdulla Rahil, Rupert Gammon, Neil Brown

Abstract:

The rapid growth of renewable energy sources and their integration into the grid have been motivated by the depletion of fossil fuels and environmental issues. Unfortunately, the grid is unable to cope with the predicted growth of renewable energy which would lead to its instability. To solve this problem, energy storage devices could be used. Electrolytic hydrogen production from an electrolyser is considered a promising option since it is a clean energy source (zero emissions). Choosing flexible operation of an electrolyser (producing hydrogen during the off-peak electricity period and stopping at other times) could bring about many benefits like reducing the cost of hydrogen and helping to balance the electric systems. This paper investigates the price of hydrogen during flexible operation compared with continuous operation, while serving the customer (hydrogen filling station) without interruption. The optimization algorithm is applied to investigate the hydrogen station in both cases (flexible and continuous operation). Three different scenarios are tested to see whether the off-peak electricity price could enhance the reduction of the hydrogen cost. These scenarios are: Standard tariff (1 tier system) during the day (assumed 12 p/kWh) while still satisfying the demand for hydrogen; using off-peak electricity at a lower price (assumed 5 p/kWh) and shutting down the electrolyser at other times; using lower price electricity at off-peak times and high price electricity at other times. This study looks at Derna city, which is located on the coast of the Mediterranean Sea (32° 46′ 0 N, 22° 38′ 0 E) with a high potential for wind resource. Hourly wind speed data which were collected over 24½ years from 1990 to 2014 were in addition to data on hourly radiation and hourly electricity demand collected over a one-year period, together with the petrol station data.

Keywords: hydrogen filling station off-peak electricity, renewable energy, off-peak electricity, electrolytic hydrogen

Procedia PDF Downloads 231

Authors: S. Ajith Kumar, Vaisakh S. Rajan

Abstract:

In external flows, vortex shedding behind the bluff bodies causes to experience unsteady loads on a large number of engineering structures, resulting in structural failure. Vortex shedding can even turn out to be disastrous like the Tacoma Bridge failure incident. We need to have control over vortex shedding to get rid of this untoward condition by reducing the unsteady forces acting on the bluff body. In circular cylinders, hydrophobic surface in an otherwise no-slip surface is found to be delaying separation and minimizes the effects of vortex shedding drastically. Flow over square cylinder stands different from this behavior as separation can takes place from either of the two corner separation points (front or rear). An attempt is made in this study to numerically elucidate the effect of hydrophobic surface in flow over a square cylinder. A 2D numerical simulation has been done to understand the effects of the slip surface on the flow past square cylinder. The details of the numerical algorithm will be presented at the time of the conference. A non-dimensional parameter, Knudsen number is defined to quantify the slip on the cylinder surface based on Maxwell’s equation. The slip surface condition of the wall affects the vorticity distribution around the cylinder and the flow separation. In the numerical analysis, we observed that the hydrophobic surface enhances the shedding frequency and damps down the amplitude of oscillations of the square cylinder. We also found that the slip has a negative effect on aerodynamic force coefficients such as the coefficient of lift (CL), coefficient of drag (CD) etc. and hence replacing the no slip surface by a hydrophobic surface can be treated as an effective drag reduction strategy and the introduction of hydrophobic surface could be utilized for reducing the vortex induced vibrations (VIV) and is found as an effective method in controlling VIV thereby controlling the structural failures.

Keywords: drag reduction, flow past square cylinder, flow control, hydrophobic surfaces, vortex shedding

Procedia PDF Downloads 373

Authors: Iffat Ara Ebu, Fahmida Islam, Mohammad Abdus Shahid Rafi, Mahfuzur Rahman, Umar Iqbal, John Ball

Abstract:

The application of multi-sensor fusion for enhanced distance estimation accuracy in dynamic environments is crucial for advanced driver assistance systems (ADAS) and autonomous vehicles. Limitations of single sensors such as cameras or radar in adverse conditions motivate the use of combined camera and radar data to improve reliability, adaptability, and object recognition. A multi-sensor fusion approach using an extended Kalman filter (EKF) is proposed to combine sensor measurements with a dynamic system model, achieving robust and accurate distance estimation. The research utilizes the Mississippi State University Autonomous Vehicular Simulator (MAVS) to create a controlled environment for data collection. Data analysis is performed using MATLAB. Qualitative (visualization of fused data vs ground truth) and quantitative metrics (RMSE, MAE) are employed for performance assessment. Initial results with simulated data demonstrate accurate distance estimation compared to individual sensors. The optimal sensor measurement noise variance and plant noise variance parameters within the EKF are identified, and the algorithm is validated with real-world data from a Chevrolet Blazer. In summary, this research demonstrates that multi-sensor fusion with an EKF significantly improves distance estimation accuracy in dynamic environments. This is supported by comprehensive evaluation metrics, with validation transitioning from simulated to real-world data, paving the way for safer and more reliable autonomous vehicle control.

Keywords: sensor fusion, EKF, MATLAB, MAVS, autonomous vehicle, ADAS

Procedia PDF Downloads 43

Authors: Saeid Gharechelou, Stuart Green, Fiona Cawkwell

Abstract:

Hedgerow has an important role for a wide range of ecological habitats, landscape, agriculture management, carbon sequestration, wood production. Hedgerow detection accurately using satellite imagery is a challenging problem in remote sensing techniques, because in the special approach it is very similar to line object like a road, from a spectral viewpoint, a hedge is very similar to a forest. Remote sensors with very high spatial resolution (VHR) recently enable the automatic detection of hedges by the acquisition of images with enough spectral and spatial resolution. Indeed, recently VHR remote sensing data provided the opportunity to detect the hedgerow as line feature but still remain difficulties in monitoring the characterization in landscape scale. In this research is used the TerraSAR-x Spotlight and Staring mode with 3-5 m resolution in wet and dry season in the test site of Fermoy County, Ireland to detect the hedgerow by acquisition time of 2014-2015. Both dual polarization of Spotlight data in HH/VV is using for detection of hedgerow. The varied method of SAR image technique with try and error way by integration of classification algorithm like texture analysis, support vector machine, k-means and random forest are using to detect hedgerow and its characterization. We are applying the Shannon entropy (ShE) and backscattering analysis in single and double bounce in polarimetric analysis for processing the object-oriented classification and finally extracting the hedgerow network. The result still is in progress and need to apply the other method as well to find the best method in study area. Finally, this research is under way to ahead to get the best result and here just present the preliminary work that polarimetric image of TSX potentially can detect the hedgerow.

Keywords: TerraSAR-X, hedgerow detection, high resolution SAR image, dual polarization, polarimetric analysis

Procedia PDF Downloads 230

Authors: Abdelrhman Elagez, Rolla Monib

Abstract:

This empirical study investigates the observed risks associated with adopting Artificial Intelligence (AI) technologies in the construction industry and proposes potential mitigation strategies. While AI has transformed several industries, the construction industry is slowly adopting advanced technologies like AI, introducing new risks that lack critical analysis in the current literature. A comprehensive literature review identified a research gap, highlighting the lack of critical analysis of risks and the need for a framework to measure and mitigate the risks of AI implementation in the construction industry. Consequently, an online survey was conducted with 24 project managers and construction professionals, possessing experience ranging from 1 to 30 years (with an average of 6.38 years), to gather industry perspectives and concerns relating to AI integration. The survey results yielded several significant findings. Firstly, respondents exhibited a moderate level of familiarity (66.67%) with AI technologies, while the industry's readiness for AI deployment and current usage rates remained low at 2.72 out of 5. Secondly, the top-ranked barriers to AI adoption were identified as lack of awareness, insufficient knowledge and skills, data quality concerns, high implementation costs, absence of prior case studies, and the uncertainty of outcomes. Thirdly, the most significant risks associated with AI use in construction were perceived to be a lack of human control (decision-making), accountability, algorithm bias, data security/privacy, and lack of legislation and regulations. Additionally, the participants acknowledged the value of factors such as education, training, organizational support, and communication in facilitating AI integration within the industry. These findings emphasize the necessity for tailored risk assessment frameworks, guidelines, and governance principles to address the identified risks and promote the responsible adoption of AI technologies in the construction sector.

Keywords: risk management, construction, artificial intelligence, technology

Procedia PDF Downloads 98

Authors: Guilherme Giacomini, Ana Luiza Menegatti Pavan, Allan Felipe Fattori Alves, Marcela de Oliveira, Fernando Antonio Bacchim Neto, José Ricardo de Arruda Miranda, Seizo Yamashita, Diana Rodrigues de Pina

Abstract:

The maxillary sinus (MS), part of the paranasal sinus complex, is one of the most enigmatic structures in modern humans. The literature has suggested that MSs function as olfaction accessories, to heat or humidify inspired air, for thermoregulation, to impart resonance to the voice and others. Thus, the real function of the MS is still uncertain. Furthermore, the MS anatomy is complex and varies from person to person. Many diseases may affect the development process of sinuses. The incidence of rhinosinusitis and other pathoses in the MS is comparatively high, so, volume analysis has clinical value. Providing volume values for MS could be helpful in evaluating the presence of any abnormality and could be used for treatment planning and evaluation of the outcome. The computed tomography (CT) has allowed a more exact assessment of this structure, which enables a quantitative analysis. However, this is not always possible in the clinical routine, and if possible, it involves much effort and/or time. Therefore, it is necessary to have a convenient, robust, and practical tool correlated with the MS volume, allowing clinical applicability. Nowadays, the available methods for MS segmentation are manual or semi-automatic. Additionally, manual methods present inter and intraindividual variability. Thus, the aim of this study was to develop an automatic tool to quantity the MS volume in CT scans of paranasal sinuses. This study was developed with ethical approval from the authors’ institutions and national review panels. The research involved 30 retrospective exams of University Hospital, Botucatu Medical School, São Paulo State University, Brazil. The tool for automatic MS quantification, developed in Matlab®, uses a hybrid method, combining different image processing techniques. For MS detection, the algorithm uses a Support Vector Machine (SVM), by features such as pixel value, spatial distribution, shape and others. The detected pixels are used as seed point for a region growing (RG) segmentation. Then, morphological operators are applied to reduce false-positive pixels, improving the segmentation accuracy. These steps are applied in all slices of CT exam, obtaining the MS volume. To evaluate the accuracy of the developed tool, the automatic method was compared with manual segmentation realized by an experienced radiologist. For comparison, we used Bland-Altman statistics, linear regression, and Jaccard similarity coefficient. From the statistical analyses for the comparison between both methods, the linear regression showed a strong association and low dispersion between variables. The Bland–Altman analyses showed no significant differences between the analyzed methods. The Jaccard similarity coefficient was > 0.90 in all exams. In conclusion, the developed tool to quantify MS volume proved to be robust, fast, and efficient, when compared with manual segmentation. Furthermore, it avoids the intra and inter-observer variations caused by manual and semi-automatic methods. As future work, the tool will be applied in clinical practice. Thus, it may be useful in the diagnosis and treatment determination of MS diseases. Providing volume values for MS could be helpful in evaluating the presence of any abnormality and could be used for treatment planning and evaluation of the outcome. The computed tomography (CT) has allowed a more exact assessment of this structure which enables a quantitative analysis. However, this is not always possible in the clinical routine, and if possible, it involves much effort and/or time. Therefore, it is necessary to have a convenient, robust and practical tool correlated with the MS volume, allowing clinical applicability. Nowadays, the available methods for MS segmentation are manual or semi-automatic. Additionally, manual methods present inter and intraindividual variability. Thus, the aim of this study was to develop an automatic tool to quantity the MS volume in CT scans of paranasal sinuses. This study was developed with ethical approval from the authors’ institutions and national review panels. The research involved 30 retrospective exams of University Hospital, Botucatu Medical School, São Paulo State University, Brazil. The tool for automatic MS quantification, developed in Matlab®, uses a hybrid method, combining different image processing techniques. For MS detection, the algorithm uses a Support Vector Machine (SVM), by features such as pixel value, spatial distribution, shape and others. The detected pixels are used as seed point for a region growing (RG) segmentation. Then, morphological operators are applied to reduce false-positive pixels, improving the segmentation accuracy. These steps are applied in all slices of CT exam, obtaining the MS volume. To evaluate the accuracy of the developed tool, the automatic method was compared with manual segmentation realized by an experienced radiologist. For comparison, we used Bland-Altman statistics, linear regression and Jaccard similarity coefficient. From the statistical analyses for the comparison between both methods, the linear regression showed a strong association and low dispersion between variables. The Bland–Altman analyses showed no significant differences between the analyzed methods. The Jaccard similarity coefficient was > 0.90 in all exams. In conclusion, the developed tool to automatically quantify MS volume proved to be robust, fast and efficient, when compared with manual segmentation. Furthermore, it avoids the intra and inter-observer variations caused by manual and semi-automatic methods. As future work, the tool will be applied in clinical practice. Thus, it may be useful in the diagnosis and treatment determination of MS diseases.

Keywords: maxillary sinus, support vector machine, region growing, volume quantification

Procedia PDF Downloads 504

Authors: Ana Neves, John Leander, Ignacio Gonzalez, Raid Karoumi

Abstract:

Systematic monitoring and inspection are needed to assess the present state of a structure and predict its future condition. If an irregularity is noticed, repair actions may take place and the adequate intervention will most probably reduce the future costs with maintenance, minimize downtime and increase safety by avoiding the failure of the structure as a whole or of one of its structural parts. For this to be possible decisions must be made at the right time, which implies using systems that can detect abnormalities in their early stage. In this sense, Structural Health Monitoring (SHM) is seen as an effective tool for improving the safety and reliability of infrastructures. This paper explores the decision-making problem in SHM regarding the maintenance of civil engineering structures. The aim is to assess the present condition of a bridge based exclusively on measurements using the suggested method in this paper, such that action is taken coherently with the information made available by the monitoring system. Artificial Neural Networks are trained and their ability to predict structural behavior is evaluated in the light of a case study where acceleration measurements are acquired from a bridge located in Stockholm, Sweden. This relatively old bridge is presently still in operation despite experiencing obvious problems already reported in previous inspections. The prediction errors provide a measure of the accuracy of the algorithm and are subjected to further investigation, which comprises concepts like clustering analysis and statistical hypothesis testing. These enable to interpret the obtained prediction errors, draw conclusions about the state of the structure and thus support decision making regarding its maintenance.

Keywords: artificial neural networks, clustering analysis, model-free damage detection, statistical hypothesis testing, structural health monitoring

Procedia PDF Downloads 208

Authors: Zainab Mutlaq Ibrahim, Mohamed Bader-El-Den, Mihaela Cocea

Abstract:

Students' textual feedback can hold unique patterns and useful information about learning process, it can hold information about advantages and disadvantages of teaching methods, assessment components, facilities, and other aspects of teaching. The results of analysing such a feedback can form a key point for institutions’ decision makers to advance and update their systems accordingly. This paper proposes a data mining framework for analysing end of unit general textual feedback using part of speech feature (PoS) with four machine learning algorithms: support vector machines, decision tree, random forest, and naive bays. The proposed framework has two tasks: first, to use the above algorithms to build an optimal model that automatically classifies the whole data set into two subsets, one subset is tailored to assessment practices (assessment related), and the other one is the non-assessment related data. Second task to use the same algorithms to build an optimal model for whole data set, and the new data subsets to automatically detect their sentiment. The significance of this paper is to compare the performance of the above four algorithms using part of speech feature to the performance of the same algorithms using n-grams feature. The paper follows Knowledge Discovery and Data Mining (KDDM) framework to construct the classification and sentiment analysis models, which is understanding the assessment domain, cleaning and pre-processing the data set, selecting and running the data mining algorithm, interpreting mined patterns, and consolidating the discovered knowledge. The results of this paper experiments show that both models which used both features performed very well regarding first task. But regarding the second task, models that used part of speech feature has underperformed in comparison with models that used unigrams and bigrams.

Keywords: assessment, part of speech, sentiment analysis, student feedback

Procedia PDF Downloads 142

Authors: Abdesselem Dakhli, Wajdi Bellil, Chokri Ben Amar

Abstract:

DNA Barcode, a short mitochondrial DNA fragment, made up of three subunits; a phosphate group, sugar and nucleic bases (A, T, C, and G). They provide good sources of information needed to classify living species. Such intuition has been confirmed by many experimental results. Species classification with DNA Barcode sequences has been studied by several researchers. The classification problem assigns unknown species to known ones by analyzing their Barcode. This task has to be supported with reliable methods and algorithms. To analyze species regions or entire genomes, it becomes necessary to use similarity sequence methods. A large set of sequences can be simultaneously compared using Multiple Sequence Alignment which is known to be NP-complete. To make this type of analysis feasible, heuristics, like progressive alignment, have been developed. Another tool for similarity search against a database of sequences is BLAST, which outputs shorter regions of high similarity between a query sequence and matched sequences in the database. However, all these methods are still computationally very expensive and require significant computational infrastructure. Our goal is to build predictive models that are highly accurate and interpretable. This method permits to avoid the complex problem of form and structure in different classes of organisms. On empirical data and their classification performances are compared with other methods. Our system consists of three phases. The first is called transformation, which is composed of three steps; Electron-Ion Interaction Pseudopotential (EIIP) for the codification of DNA Barcodes, Fourier Transform and Power Spectrum Signal Processing. The second is called approximation, which is empowered by the use of Multi Llibrary Wavelet Neural Networks (MLWNN).The third is called the classification of DNA Barcodes, which is realized by applying the algorithm of hierarchical classification.

Keywords: DNA barcode, electron-ion interaction pseudopotential, Multi Library Wavelet Neural Networks (MLWNN)

Procedia PDF Downloads 317

Authors: Kashif Habib, Zeeshan Ayyub

Abstract:

ZnO voltage dependent resistor was widely used as components of the electrical system for over-voltage protection. It has a wide application prospect in superconducting energy-removal, generator de-excitation, overvoltage protection of electrical & electronics equipment. At present, the research for the application of ZnO voltage dependent resistor stop, it uses just in the field of its nonlinear voltage current characteristic and overvoltage protection areas. There is no further study over the over-voltage breakdown characteristics, such as the combustion phenomena and the measure of the voltage/current when it breakdown, and the affect to its surrounding equipment. It is also a blind spot in its application. So, when we do the feature test of ZnO voltage dependent resistor, we need to design a reasonable test power supply, making the terminal voltage keep for sine wave, simulating the real use of PF voltage in power supply conditions. We put forward the solutions of using inverter to generate a controllable power. The paper mainly focuses on the breakdown characteristic test power supply of nonlinear ZnO voltage dependent resistor. According to the current mature switching power supply technology, we proposed power control system using the inverter as the core. The power mainly realize the sin-voltage output on the condition of three-phase PF-AC input, and 3 control modes (RMS, Peak, Average) of the current output. We choose TMS320F2812M as the control part of the hardware platform. It is used to convert the power from three-phase to a controlled single-phase sin-voltage through a rectifier, filter, and inverter. Design controller produce SPWM, to get the controlled voltage source via appropriate multi-loop control strategy, while execute data acquisition and display, system protection, start logic control, etc. The TMS320F2812M is able to complete the multi-loop control quickly and can be a good completion of the inverter output control.

Keywords: ZnO, multi-loop control, SPWM, non-linear load

Procedia PDF Downloads 325

Authors: Waqqas-ur-Rehman Butt, Martin Servin, Marion Pause

Abstract:

In recent years, object detection has gained much attention and very encouraging research area in the field of computer vision. The robust object boundaries detection in an image is demanded in numerous applications of human computer interaction and automated surveillance systems. Many methods and approaches have been developed for automatic object detection in various fields, such as automotive, quality control management and environmental services. Inappropriately, to the best of our knowledge, object detection under illumination with shadow consideration has not been well solved yet. Furthermore, this problem is also one of the major hurdles to keeping an object detection method from the practical applications. This paper presents an approach to automatic object detection in images under non-standardized environmental conditions. A key challenge is how to detect the object, particularly under uneven illumination conditions. Image capturing conditions the algorithms need to consider a variety of possible environmental factors as the colour information, lightening and shadows varies from image to image. Existing methods mostly failed to produce the appropriate result due to variation in colour information, lightening effects, threshold specifications, histogram dependencies and colour ranges. To overcome these limitations we propose an object detection algorithm, with pre-processing methods, to reduce the interference caused by shadow and illumination effects without fixed parameters. We use the Y CrCb colour model without any specific colour ranges and predefined threshold values. The segmented object regions are further classified using morphological operations (Erosion and Dilation) and contours. Proposed approach applied on a large image data set acquired under various environmental conditions for wood stack detection. Experiments show the promising result of the proposed approach in comparison with existing methods.

Keywords: image processing, illumination equalization, shadow filtering, object detection

Procedia PDF Downloads 216

Authors: Xueming Xu, Jiahao Liu, Jichao Li, Kewei Yang, Minghao Li, Bingfeng Ge

Abstract:

A combat system-of-systems (CSoS) comprises various types of functional combat entities that interact to meet corresponding task requirements in the present and future. Enhancing the resilience of CSoS holds significant military value in optimizing the operational planning process, improving military survivability, and ensuring the successful completion of operational tasks. Accordingly, this research proposes an integrated framework called CSoS resilience enhancement (CSoSRE) to enhance the resilience of CSoS from a recovery perspective. Specifically, this research presents a two-phase resilience optimization model to define a resilience optimization objective for CSoS. This model considers not only task baseline, recovery cost, and recovery time limit but also the characteristics of emergency recovery and comprehensive recovery. Moreover, the research extends it from the deterministic case to the stochastic case to describe the uncertainty in the recovery process. Based on this, a resilience-oriented recovery optimization method based on deep reinforcement learning (RRODRL) is proposed to determine a set of entities requiring restoration and their recovery sequence, thereby enhancing the resilience of CSoS. This method improves the deep Q-learning algorithm by designing a discount factor that adapts to changes in CSoS state at different phases, simultaneously considering the network’s structural and functional characteristics within CSoS. Finally, extensive experiments are conducted to test the feasibility, effectiveness and superiority of the proposed framework. The obtained results offer useful insights for guiding operational recovery activity and designing a more resilient CSoS.

Keywords: combat system-of-systems, resilience optimization model, recovery optimization method, deep reinforcement learning, certainty and uncertainty

Procedia PDF Downloads 16

Authors: Li-Ting Huang, Yu-Hsiang Shen, Cing-Ciao Ke, Sheng-Pin Tseng, Fan-Pin Tseng, Yu-Ching Ni, Chia-Yu Lin

Abstract:

Chest radiography is the most common technique for the diagnosis and follow-up of pulmonary diseases. However, the lesions superimposed with normal structures are difficult to be detected in chest radiography. Chest tomosynthesis is a relatively new technique to obtain 3D section images from a set of low-dose projections acquired over a limited angular range. However, there are some limitations with chest tomosynthesis. Patients undergoing tomosynthesis have to be able to hold their breath firmly for 10 seconds. A digital tomosynthesis system with advanced reconstruction algorithm and high-stability motion mechanism was developed by our research group. The potential for the system to perform a bidirectional chest scan within 10 seconds is expected. The purpose of this study is to realize the influences of the scan speed on the image quality for our digital tomosynthesis system. The major factors that lead image blurring are the motion of the X-ray source and the patient. For the fore one, an experiment of imaging a chest phantom with three different scan speeds, which are 6 cm/s, 8 cm/s, and 15 cm/s, was proceeded to understand the scan speed influences on the image quality. For the rear factor, a normal SD (Sprague-Dawley) rat was imaged with it alive and sacrificed to assess the impact on the image quality due to breath motion. In both experiments, the profile of the ROIs (region of interest) and the CNRs (contrast-to-noise ratio) of the ROIs to the normal tissue of the reconstructed images was examined to realize the degradations of the qualities of the images. The preliminary results show that no obvious degradation of the image quality was observed with increasing scan speed, possibly due to the advanced designs for the hardware and software of the system. It implies that higher speed (15 cm/s) than that of the commercialized tomosynthesis system (12 cm/s) for the proposed system is achieved, and therefore a complete chest scan within 10 seconds is expected.

Keywords: chest radiography, digital tomosynthesis, image quality, scan speed

Procedia PDF Downloads 331

Authors: H. Fuad Usman, M. Rafay Khan Sial, Shahzaib Hamid

Abstract:

The trend of renewable energy resources has been amplified due to global warming and other environmental related complications in the 21st century. Recent research has very much emphasized on the generation of electrical power through renewable resources like solar, wind, hydro, geothermal, etc. The use of the photovoltaic cell has become very public as it is very useful for the domestic and commercial purpose overall the world. Although a single cell gives the low voltage output but connecting a number of cells in a series formed a complete module of the photovoltaic cells, it is becoming a financial investment as the use of it fetching popular. This also reduced the prices of the photovoltaic cell which gives the customers a confident of using this source for their electrical use. Photovoltaic cell gives the MPPT at single specific point of operation at a given temperature and level of solar intensity received at a given surface whereas the focal point changes over a large range depending upon the manufacturing factor, temperature conditions, intensity for insolation, instantaneous conditions for shading and aging factor for the photovoltaic cells. Two improved algorithms have been proposed in this article for the MPPT. The widely used algorithms are the ‘Incremental Conductance’ and ‘Perturb and Observe’ algorithms. To extract the maximum power from the source to the load, the duty cycle of the convertor will be effectively controlled. After assessing the previous techniques, this paper presents the improved and reformed idea of harvesting maximum power point from the photovoltaic cells. A thoroughly go through of the previous ideas has been observed before constructing the improvement in the traditional technique of MPP. Each technique has its own importance and boundaries at various weather conditions. An improved technique of implementing the use of both ‘Perturb and Observe’ and ‘Incremental Conductance’ is introduced.

Keywords: duty cycle, MPPT (Maximum Power Point Tracking), perturb and observe (P&O), photovoltaic module

Procedia PDF Downloads 176

Authors: Eman W. Boghdady, A. R. Shehata, M. A. Azem

Abstract:

Database (DB) security demands permitting authorized users and prohibiting non-authorized users and intruders actions on the DB and the objects inside it. Organizations that are running successfully demand the confidentiality of their DBs. They do not allow the unauthorized access to their data/information. They also demand the assurance that their data is protected against any malicious or accidental modification. DB protection and confidentiality are the security concerns. There are four types of controls to obtain the DB protection, those include: access control, information flow control, inference control, and cryptographic. The cryptographic control is considered as the backbone for DB security, it secures the DB by encryption during storage and communications. Current cryptographic techniques are classified into two types: traditional classical cryptography using standard algorithms (DES, AES, IDEA, etc.) and chaos cryptography using continuous (Chau, Rossler, Lorenz, etc.) or discreet (Logistics, Henon, etc.) algorithms. The important characteristics of chaos are its extreme sensitivity to initial conditions of the system. In this paper, DB-security systems based on chaotic algorithms are described. The Pseudo Random Numbers Generators (PRNGs) from the different chaotic algorithms are implemented using Matlab and their statistical properties are evaluated using NIST and other statistical test-suits. Then, these algorithms are used to secure conventional DB (plaintext), where the statistical properties of the ciphertext are also tested. To increase the complexity of the PRNGs and to let pass all the NIST statistical tests, we propose two hybrid PRNGs: one based on two chaotic Logistic maps and another based on two chaotic Henon maps, where each chaotic algorithm is running side-by-side and starting from random independent initial conditions and parameters (encryption keys). The resulted hybrid PRNGs passed the NIST statistical test suit.

Keywords: algorithms and data structure, DB security, encryption, chaotic algorithms, Matlab, NIST

Procedia PDF Downloads 265

Authors: Babak Bahri, Fatemeh Yassaee Meybodi, Changiz Eslahchi

Abstract:

In the pursuit of effective cancer therapies, the exploration of combinatorial drug regimens is crucial to leverage synergistic interactions between drugs, thereby improving treatment efficacy and overcoming drug resistance. However, identifying synergistic drug pairs poses challenges due to the vast combinatorial space and limitations of experimental approaches. This study introduces ClusterSyn, a machine learning (ML)-powered framework for classifying anti-cancer drug synergy scores. ClusterSyn employs a two-step approach involving drug clustering and synergy score prediction using a fully connected deep neural network. For each cell line in the training dataset, a drug graph is constructed, with nodes representing drugs and edge weights denoting synergy scores between drug pairs. Drugs are clustered using the Markov clustering (MCL) algorithm, and vectors representing the similarity of drug pairs to each cluster are input into the deep neural network for synergy score prediction (synergy or antagonism). Clustering results demonstrate effective grouping of drugs based on synergy scores, aligning similar synergy profiles. Subsequently, neural network predictions and synergy scores of the two drugs on others within their clusters are used to predict the synergy score of the considered drug pair. This approach facilitates comparative analysis with clustering and regression-based methods, revealing the superior performance of ClusterSyn over state-of-the-art methods like DeepSynergy and DeepDDS on diverse datasets such as Oniel and Almanac. The results highlight the remarkable potential of ClusterSyn as a versatile tool for predicting anti-cancer drug synergy scores.

Keywords: drug synergy, clustering, prediction, machine learning., deep learning

Procedia PDF Downloads 79

Authors: Souvik Phadikar, Nidul Sinha, Rajdeep Ghosh

Abstract:

In brainwaves research using electroencephalogram (EEG) signals, finding the most relevant and effective feature set for identification of activities in the human brain is a big challenge till today because of the random nature of the signals. The feature extraction method is a key issue to solve this problem. Finding those features that prove to give distinctive pictures for different activities and similar for the same activities is very difficult, especially for the number of activities. The performance of a classifier accuracy depends on this quality of feature set. Further, more number of features result in high computational complexity and less number of features compromise with the lower performance. In this paper, a novel idea of the selection of optimal feature set using a heuristically optimized deep autoencoder is presented. Using various feature extraction methods, a vast number of features are extracted from the EEG signals and fed to the autoencoder deep neural network. The autoencoder encodes the input features into a small set of codes. To avoid the gradient vanish problem and normalization of the dataset, a meta-heuristic search algorithm is used to minimize the mean square error (MSE) between encoder input and decoder output. To reduce the feature set into a smaller one, 4 hidden layers are considered in the autoencoder network; hence it is called Heuristically Optimized Deep Autoencoder (HO-DAE). In this method, no features are rejected; all the features are combined into the response of responses of the hidden layer. The results reveal that higher accuracy can be achieved using optimal reduced features. The proposed HO-DAE is also compared with the regular autoencoder to test the performance of both. The performance of the proposed method is validated and compared with the other two methods recently reported in the literature, which reveals that the proposed method is far better than the other two methods in terms of classification accuracy.

Keywords: autoencoder, brainwave signal analysis, electroencephalogram, feature extraction, feature selection, optimization

Procedia PDF Downloads 114