Search results for: content classification

Authors: Danilo Habermann, Aleksander Medella, Carla Cremon, Yusef Caceres

Abstract:

This paper aims to analyze the ability of 2d point clouds features to classify different models of helicopters using radars. This method does not need to estimate the blade length, the number of blades of helicopters, and the period of their micro-Doppler signatures. It is also not necessary to generate spectrograms (or any other image based on time and frequency domain). This work transforms a radar return signal into a 2D point cloud and extracts features of it. Three classifiers are used to distinguish 9 different helicopter models in order to analyze the performance of the features used in this work. The high accuracy obtained with each of the classifiers demonstrates that the 2D point clouds features are very useful for classifying helicopters from radar signal.

Keywords: helicopter classification, point clouds features, radar, supervised classifiers

Procedia PDF Downloads 214

Authors: Radwa Mabrook

Abstract:

Virtual Reality (VR) content creation is a complex and an expensive process, which requires multi-disciplinary teams of content creators. Grant schemes from technology companies help media organisations to explore the VR potential in journalism and factual storytelling. Media organisations try to do as much as they can in-house, but they may outsource due to time constraints and skill availability. Journalists, game developers, sound designers and creative artists work together and bring in new cultures of work. This study explores the collaborative experimental nature of VR content creation, through tracing every actor involved in the process and examining their perceptions of the VR work. The study builds on Actor Network Theory (ANT), which decomposes phenomena into their basic elements and traces the interrelations among them. Therefore, the researcher conducted 22 semi-structured interviews with VR content creators between November 2017 and April 2018. Purposive and snowball sampling techniques allowed the researcher to recruit fact-based VR content creators from production studios and media organisations, as well as freelancers. Interviews lasted up to three hours, and they were a mix of Skype calls and in-person interviews. Participants consented for their interviews to be recorded, and for their names to be revealed in the study. The researcher coded interviews’ transcripts in Nvivo software, looking for key themes that correspond with the research questions. The study revealed that VR content creators must be adaptive to change, open to learn and comfortable with mistakes. The VR content creation process is very iterative because VR has no established work flow or visual grammar. Multi-disciplinary VR team members often speak different languages making it hard to communicate. However, adaptive content creators perceive VR work as a fun experience and an opportunity to learn. The traditional sense of competition and the strive for information exclusivity are now replaced by a strong drive for knowledge sharing. VR content creators are open to share their methods of work and their experiences. They target to build a collaborative network that aims to harness VR technology for journalism and factual storytelling. Indeed, VR is instilling collaborative and experimental cultures in journalism.

Keywords: collaborative culture, content creation, experimental culture, virtual reality

Procedia PDF Downloads 120

Authors: Danielle Shackley, Yetunde Folajimi

Abstract:

As more people turn to the internet seeking health-related information, there is more risk of finding false, inaccurate, or dangerous information. Sentiment analysis is a natural language processing technique that assigns polarity scores to text, ranging from positive, neutral, and negative. In this research, we evaluate the weight of a sentiment analysis feature added to fake health news classification models. The dataset consists of existing reliably labeled health article headlines that were supplemented with health information collected about COVID-19 from social media sources. We started with data preprocessing and tested out various vectorization methods such as Count and TFIDF vectorization. We implemented 3 Naive Bayes classifier models, including Bernoulli, Multinomial, and Complement. To test the weight of the sentiment analysis feature on the dataset, we created benchmark Naive Bayes classification models without sentiment analysis, and those same models were reproduced, and the feature was added. We evaluated using the precision and accuracy scores. The Bernoulli initial model performed with 90% precision and 75.2% accuracy, while the model supplemented with sentiment labels performed with 90.4% precision and stayed constant at 75.2% accuracy. Our results show that the addition of sentiment analysis did not improve model precision by a wide margin; while there was no evidence of improvement in accuracy, we had a 1.9% improvement margin of the precision score with the Complement model. Future expansion of this work could include replicating the experiment process and substituting the Naive Bayes for a deep learning neural network model.

Keywords: sentiment analysis, Naive Bayes model, natural language processing, topic analysis, fake health news classification model

Procedia PDF Downloads 87

Authors: Madhuri Kanala, Geoffrey Savage

Abstract:

Worldwide CO₂ levels are slowly rising, and this may have effects on the growth and nutritional composition of many food plants. The production of secondary metabolites such as oxalates has not been investigated in depth. The oxalate content of many food plants are known to have adverse nutritional effects on humans and reduction in the oxalate contents of food plants is a very positive move. Recent studies had shown that the oxalate content of the leaves of spinach and silver beet reduced when the plants were grown in an environment where CO₂ was increased. The response of amaranth and strawberry spinach leaves to changes in the high CO₂ environment have not been understood though it is known that the plants do contain appreciable oxalate contents. A study was conducted where amaranth and strawberry spinach plants were grown in identical plant growth chambers with the same environmental conditions except that one chamber was supplied with ambient air (CO₂ 405 ppm) while the other chamber had the CO₂ level increased to 650 ppm. The total and soluble oxalate content of the leaves of raw and cooked amaranth and strawberry spinach were determined by HPLC and calcium levels were determined using ICP following 6 weeks of growth. The total oxalate content of the fresh leaves of amaranth and strawberry spinach were reduced by 29.5 % and 24.6% respectively in the leaves of the plants grown in increased CO₂ conditions compared to ambient levels. The soluble oxalate content of amaranth leaves grown under ambient and increased CO₂ conditions were future reduced by 42% and 26.8% respectively following cooking as the soluble oxalate was leached into the cooking water and discarded. The reduction of the oxalate and calcium levels of raw and cooked amaranth and strawberry spinach leaves following an increase in CO₂ content in the air is an interesting positive response to an otherwise significant environmental problem.

Keywords: amaranth, calcium oxalate, enriched CO₂, oxalates, strawberry spinach

Procedia PDF Downloads 182

Authors: R. Sadeghzadegan, S. A. Naeini, A. Mirzaii

Abstract:

In this article, the results of a series of carefully conducted laboratory test program were represented to determine the small strain shear modulus of sand mixed with a range of kaolinite including zero to 30%. This was experimentally achieved using a triaxial cell equipped with bender element. Results indicate that small shear modulus tends to increase, while clay content decreases and effective confining pressure increases. The exponent of stress in the power model regression analysis was not sensitive to the amount of clay content for all sand clay mixtures, while coefficient A was directly affected by change in clay content.

Keywords: small shear modulus, bender element test, plastic fines, sand

Procedia PDF Downloads 463

Authors: S. Noreen, S. Ahmad

Abstract:

Salinity is one of the most important environmental factors that limit the production of crop plants to the greatest proportion than any other ones. Salt-induced changes in growth, pigment concentration, water status, malondialdehydes (MDA) and H₂O₂ content, enzymatic and non-enzymatic antioxidants, Na⁺, K⁺ content and yield attributes were examined in the glasshouse on ten pea (Pisum Sativum L.) accessions, namely ‘13240’, ‘18302’, ‘19666’, ‘19700’, ‘19776’, ‘19785’, ‘19788’, ‘20153’, ‘20155’, ‘26719’ were subjected to non-stress (0 mM NaCl) and salt stress (100 mM and150 mM NaCl) in pots containing sand medium. The results showed that salt stress at level150 mM substantially reduced biomass production, leaf water status, pigment concentration (chlorophyll ‘a’, ‘b’, ‘carotenoid content’ total chlorophyll), K⁺ content, quantum yield and yield attributes as compared to plants treated with 100 mM NaCl. Antioxidant enzymes, Catalase (CAT), Peroxidase (POD), Superoxide dismutase (SOD) and Ascorbate peroxidase (APX), proline content, total soluble protein, total amino acids, Malondialdehyde content (MDA), Hydrogen peroxide (H₂O₂) content and Na⁺ uptake markedly enhanced due to the influence of salt stress. On the basis of analyses (expressed as percent of control), of 10 accessions of pea plant, two were ranked as salt tolerant namely (‘19666’, ‘20153’), four were moderately tolerant namely (‘19700’, ‘19776’, ‘19785’, ‘20155’), and three were salt sensitive namely (‘13240’, ‘18302’, ‘26719’) at 150 mM NaCl level.

Keywords: antioxidant enzymes, ion uptake, pigment concentration, salt stress, yield attributes

Procedia PDF Downloads 99

Authors: Getenet Aseged Zeleke

Abstract:

The aim of this project was to study the synthesis of activated carbon from low-cost animal beef and the characterization of the product obtained. The bone was carbonized in an inert atmosphere at three different temperatures (500°C, 700oC and 900°C) in an electric furnace, followed by activation with hydrochloric acid. The activated animal bone charcoals obtained were characterized by using scanning electron microscopy (SEM)to observe the effect of activation compared to the unactivated bone charcoal. The following parameters were also determined: ash content, moisture content, volatile content, fixed carbon, pH, pore volume and bulk (apparent) density. The characterization result showed that the activated bone charcoal has good properties and is compared favorably with other reference activated carbons.

Keywords: bones, carbonization, activation, characterization, activated carbon

Procedia PDF Downloads 69

Authors: S. Thaiudom, W. Toommuangpak

Abstract:

Yoghurt ice cream (YIC) containing prebiotics and probiotics seems to be much more recognized among consumers who concern for their health. Not only can it be a benefit on consumers’ health but also its taste and freshness provide people easily accept. However, the survival of such probiotic especially Bifidobacterium longum, found in human gastrointestinal tract and to be benefit to human gut, was still needed to study in the severe condition as whipping and freezing in ice cream process. Low and full-fat yoghurt ice cream containing 2 and 10% (w/w) fat content (LYIC and FYIC), respectively was produced by mixing 20% yoghurt containing B. longum into milk ice cream mix. Fructo-oligosaccharides (FOS) or galacto-oligosaccharides (GOS) at 0, 1, and 2% (w/w) were separately used as prebiotic in order to improve the survival of B. longum. Survival of this bacteria as a function of ice cream storage time and ice cream properties were investigated. The results showed that prebiotic; especially FOS could improve viable count of B. longum. The more concentration of prebiotic used, the more is the survival of B. Longum. These prebiotics could prolong the survival of B. longum up to 60 days, and the amount of survival number was still in the recommended level (106 cfu per gram). Fat content and prebiotic did not significantly affect the total acidity and the overrun of all samples, but an increase of fat content significantly increased the fat particle size which might be because of partial coalescence found in FYIC rather than in LYIC. However, addition of GOS or FOS could reduce the fat particle size, especially in FYIC. GOS seemed to reduce the hardness of YIC rather than FOS. High fat content (10% fat) significantly influenced on lowering the melting rate of YIC better than 2% fat content due to the 3-dimension networks of fat partial coalescence theoretically occurring more in FYIC than in LYIC. However, FOS seemed to retard the melting rate of ice cream better than GOS. In conclusion, GOS and FOS in YIC with different fat content can enhance the survival of B. longum and affect physical and chemical properties of such yoghurt ice cream.

Keywords: Bifidobacterium longum, prebiotic, survival, yoghurt ice cream

Procedia PDF Downloads 147

Authors: Tomas Linas Šepetys

Abstract:

In online content-sharing service platforms, the ability of creators to restrict illicit use of audiovisual creative works has effectively been abolished, largely due to specific infrastructure where a huge volume of copyrighted audiovisual content can be made available to the public. The European Union legislator has attempted to strengthen the positions of creators in the realm of online content-sharing services. Article 17 of the new Digital Single Market Directive considers online content-sharing service providers to carry out acts of communication to the public of any creative content uploaded to their platforms by users and posits requirements to obtain licensing agreements. While such regulation intends to assert authors‘ ability to effectively control the dissemination of their creative works, it also creates threats of parody content overblocking through automated content monitoring. Such potentially paradoxical outcome of the efforts of the EU legislator to deliver economic safeguards for the creators in the online content-sharing service platforms leads to presume lack of informity on legislator‘s part regarding creative works‘ economic exploitation opportunities provided to creators in the online content-sharing infrastructure. Analysis conducted in this scientific research discloses that the aforementioned irregularities of parody and other creative content dissemination are caused by EU legislators‘ lack of assessment of value extraction conditions for parody creators in the online content-sharing service platforms. Historical and modeling research method application reveals the existence of two creative content dissemination models and their unique mechanisms of commercial value creation. Obligations to obtain licenses and liability over creative content uploaded to their platforms by users set in Article 17 of the Digital Single Market Directive represent technological replication of the proprietary dissemination model where the creator is able to restrict access to creative content apart from licensed retail channels. The online content-sharing service platforms represent an open dissemination model where the economic potential of creative content is based on the infrastructure of unrestricted access by users and partnership with advertising services offered by the platform. Balanced modeling of proprietary dissemination models in such infrastructure requires not only automated content monitoring measures but also additional regulatory monitoring solutions to separate parody and other types of creative content. An example of the Digital Single Market Directive proves that regulation can dictate not only the technological establishment of a proprietary dissemination model but also a partial reduction of the open dissemination model and cause a disbalance between the economic interests of creators relying on such models. The results of this scientific research conclude an informative role of the creative works dissemination model in the EU copyright law modernization process. A thorough understanding of the commercial prospects of the open dissemination model intrinsic to the online content-sharing service platform structure requires and encourages EU legislators to regulate safeguards for parody content dissemination. Implementing such safeguards would result in a common application of proprietary and open dissemination models in the online content-sharing service platforms and balanced protection of creators‘ economic interests explicitly based on those creative content dissemination models.

Keywords: copyright law, creative works dissemination model, digital single market directive, online content-sharing services

Procedia PDF Downloads 68

Authors: Rosa Figueroa, Christopher Flores

Abstract:

Text categorization is the problem of assigning a new document to a set of predetermined categories, on the basis of a training set of free-text data that contains documents whose category membership is known. To train a classification model, it is necessary to extract characteristics in the form of tokens that facilitate the learning and classification process. In text categorization, the feature extraction process involves the use of word sequences also known as N-grams. In general, it is expected that documents belonging to the same category share similar features. The Smith-Waterman (SW) algorithm is a dynamic programming algorithm that performs a local sequence alignment in order to determine similar regions between two strings or protein sequences. This work explores the use of SW algorithm as an alternative to feature extraction in text categorization. The dataset used for this purpose, contains 2,610 annotated documents with the classes Obese/Non-Obese. This dataset was represented in a matrix form using the Bag of Word approach. The score selected to represent the occurrence of the tokens in each document was the term frequency-inverse document frequency (TF-IDF). In order to extract features for classification, four experiments were conducted: the first experiment used SW to extract features, the second one used unigrams (single word), the third one used bigrams (two word sequence) and the last experiment used a combination of unigrams and bigrams to extract features for classification. To test the effectiveness of the extracted feature set for the four experiments, a Support Vector Machine (SVM) classifier was tuned using 20% of the dataset. The remaining 80% of the dataset together with 5-Fold Cross Validation were used to evaluate and compare the performance of the four experiments of feature extraction. Results from the tuning process suggest that SW performs better than the N-gram based feature extraction. These results were confirmed by using the remaining 80% of the dataset, where SW performed the best (accuracy = 97.10%, weighted average F-measure = 97.07%). The second best was obtained by the combination of unigrams-bigrams (accuracy = 96.04, weighted average F-measure = 95.97) closely followed by the bigrams (accuracy = 94.56%, weighted average F-measure = 94.46%) and finally unigrams (accuracy = 92.96%, weighted average F-measure = 92.90%).

Keywords: comorbidities, machine learning, obesity, Smith-Waterman algorithm

Procedia PDF Downloads 289

Authors: H. Abas Nejad, A. R. Teymoori

Abstract:

Facial expression recognition is one of the open problems in computer vision. Robust neutral face recognition in real time is a major challenge for various supervised learning based facial expression recognition methods. This is due to the fact that supervised methods cannot accommodate all appearance variability across the faces with respect to race, pose, lighting, facial biases, etc. in the limited amount of training data. Moreover, processing each and every frame to classify emotions is not required, as the user stays neutral for the majority of the time in usual applications like video chat or photo album/web browsing. Detecting neutral state at an early stage, thereby bypassing those frames from emotion classification would save the computational power. In this work, we propose a light-weight neutral vs. emotion classification engine, which acts as a preprocessor to the traditional supervised emotion classification approaches. It dynamically learns neutral appearance at Key Emotion (KE) points using a textural statistical model, constructed by a set of reference neutral frames for each user. The proposed method is made robust to various types of user head motions by accounting for affine distortions based on a textural statistical model. Robustness to dynamic shift of KE points is achieved by evaluating the similarities on a subset of neighborhood patches around each KE point using the prior information regarding the directionality of specific facial action units acting on the respective KE point. The proposed method, as a result, improves ER accuracy and simultaneously reduces the computational complexity of ER system, as validated on multiple databases.

Keywords: neutral vs. emotion classification, Constrained Local Model, procrustes analysis, Local Binary Pattern Histogram, statistical model

Procedia PDF Downloads 332

Authors: Ceren Kaya, Okan Erkaymaz, Orhan Ayar, Mahmut Özer

Abstract:

Diabetes Mellitus (Diabetes) is a disease based on insulin hormone disorders and causes high blood glucose. Clinical findings determine that diabetes can be diagnosed by electrophysiological signals obtained from the vital organs. 'Diabetic Retinopathy' is one of the most common eye diseases resulting on diabetes and it is the leading cause of vision loss due to structural alteration of the retinal layer vessels. In this study, features of horizontal and vertical Video-Oculography (VOG) signals have been used to classify non-proliferative and proliferative diabetic retinopathy disease. Twenty-five features are acquired by using discrete wavelet transform with VOG signals which are taken from 21 subjects. Two models, based on multi-layer perceptron and radial basis function, are recommended in the diagnosis of Diabetic Retinopathy. The proposed models also can detect level of the disease. We show comparative classification performance of the proposed models. Our results show that proposed the RBF model (100%) results in better classification performance than the MLP model (94%).

Keywords: diabetic retinopathy, discrete wavelet transform, multi-layer perceptron, radial basis function, video-oculography (VOG)

Procedia PDF Downloads 250

Authors: Mustafa Alhamdi

Abstract:

Industrial application to classify gamma rays and neutron events is investigated in this study using deep machine learning. The identification using a convolutional neural network and recursive neural network showed a significant improvement in predication accuracy in a variety of applications. The ability to identify the isotope type and activity from spectral information depends on feature extraction methods, followed by classification. The features extracted from the spectrum profiles try to find patterns and relationships to present the actual spectrum energy in low dimensional space. Increasing the level of separation between classes in feature space improves the possibility to enhance classification accuracy. The nonlinear nature to extract features by neural network contains a variety of transformation and mathematical optimization, while principal component analysis depends on linear transformations to extract features and subsequently improve the classification accuracy. In this paper, the isotope spectrum information has been preprocessed by finding the frequencies components relative to time and using them as a training dataset. Fourier transform implementation to extract frequencies component has been optimized by a suitable windowing function. Training and validation samples of different isotope profiles interacted with CdTe crystal have been simulated using Geant4. The readout electronic noise has been simulated by optimizing the mean and variance of normal distribution. Ensemble learning by combing voting of many models managed to improve the classification accuracy of neural networks. The ability to discriminate gamma and neutron events in a single predication approach using deep machine learning has shown high accuracy using deep learning. The paper findings show the ability to improve the classification accuracy by applying the spectrogram preprocessing stage to the gamma and neutron spectrums of different isotopes. Tuning deep machine learning models by hyperparameter optimization of neural network models enhanced the separation in the latent space and provided the ability to extend the number of detected isotopes in the training database. Ensemble learning contributed significantly to improve the final prediction.

Keywords: machine learning, nuclear physics, Monte Carlo simulation, noise estimation, feature extraction, classification

Procedia PDF Downloads 140

Authors: Rita Riekstina-Dolge, Zanda Kruma, Daina Karklina, Fredijs Dimins

Abstract:

Apple juice is the main raw material for cider production. In this study apple juices obtained from 14 dessert and crab variety apples grown in Latvia were investigated. For all samples soluble solids, titratable acidity, pH and sugar content were determined. Crab apples produce more dry matter, total sugar and acid content compared to the dessert apples but it depends on the apple variety. Total sugar content of crab apple juices was 1.3 to 1.8 times larger than in dessert apple juices. Titratable acidity of dessert apple juices is in the range of 4.1g L-1 to 10.83g L-1 and in crab apple juices titratable acidity is from 7.87g L-1 to 19.6g L-1. Fructose was detected as the main sugar whereas glucose level varied depending on the variety. The highest titratable acidity and content of sugars was detected in ‘Cornelia’ apples juice.

Keywords: apple juice, hierarchical cluster analysis, sugars, titratable acidity

Procedia PDF Downloads 234

Authors: Paraskevi Theodorou, Apostolos Meliones

Abstract:

In the context of the development process of two assistive navigation mobile apps for blind and visually impaired people (BVI) an extensive qualitative analysis of the requirements of potential users has been conducted. The analysis was based on interviews with BVIs and aimed to elicit not only their needs with respect to autonomous navigation but also their preferences on specific features of the apps under development. The elicited requirements were structured into four main categories, namely, requirements concerning the capabilities, functionality and usability of the apps, as well as compatibility requirements with respect to other apps and services. The main categories were then further divided into nine sub-categories. This classification, along with its content, aims to become a useful tool for the researcher or the developer who is involved in the development of digital services for BVI.

Keywords: accessibility, assistive mobile apps, blind and visually impaired people, user requirements analysis

Procedia PDF Downloads 116

Authors: Yoshimoto Kurihara, Tad Gonsalves

Abstract:

Currently, in the field of object posture estimation, there is research on estimating the position and angle of an object by storing a 3D model of the object to be estimated in advance in a computer and matching it with the model. However, in this research, we have succeeded in creating a module that is much simpler, smaller in scale, and faster in operation. Our 6D pose estimation model consists of two different networks – a classification network and a regression network. From a single RGB image, the trained model estimates the class of the object in the image, the coordinates of the object, and its rotation angle in 3D space. In addition, we compared the estimation accuracy of each camera position, i.e., the angle from which the object was captured. The highest accuracy was recorded when the camera position was 75°, the accuracy of the classification was about 87.3%, and that of regression was about 98.9%.

Keywords: 6D posture estimation, image recognition, deep learning, AlexNet

Procedia PDF Downloads 146

Authors: I Lamri, L Arabet, M. Hidjeb

Abstract:

The direct shear test was used to determine the shear strength parameters C and Ø of a series of samples with different cement content. Samples stabilized with a certain percentage of cement showed a substantial gain in compressive strength and a significant increase in shear strength parameters. C and Ø. The laboratory equipment used in UCS tests consisted of a conventional 102mm diameter sample triaxial loading machine. Beyond 4% cement content a very important increase in shear strength was observed. It can be deduced from a comparative study of shear strength of soil samples with 4%, 7%, and 10% cement with sample containing 2 %, that the sample with a 4% cement content showed 90% increase in shear strength while those with 7% and 10% showed an increase of around 13 and 21 fold.

Keywords: cement, compression strength, shear stress, cohesion, angle of internal friction

Procedia PDF Downloads 477

Authors: Huei Lin Lee, Wee Sim Choo

Abstract:

The effects of blanching, boiling and steaming on the ascorbic acid content, total phenolic content and colour in cauliflower (Brassica oleraceavar. Botrytis) was investigated. It was found that blanching was the best thermal processing to be applied on cauliflower compared to boiling and steaming processes. Blanching and steaming processes on cauliflower retained most of the ascorbic acid content (AAC) compared to those of boiling. As for the total phenolic content (TPC), blanching process retained a higher TPC in cauliflower compared to those of boiling and steaming processes. There were no significant differences between the TPC of boiled and steamed cauliflowers. As for the colour measurement, there were no significant differences in the colour of the cauliflower at different lead time (after processing to the point of consumption) of 30 minutes interval up to 3 hours but there were slight variations in L*, a*, and b* values among the thermal processed cauliflowers (blanched, boiled and steamed). The cauliflowers in this study were found to give a desirable white colour (L* value in the range of 77-83) in all the three thermal processes (blanching, boiling and steaming). There was no significant difference on the effect of lead time (30-minutes interval up to 3 hours) in raw and all the three thermal processed (blanched, boiled and steamed) cauliflowers.

Keywords: ascorbic acid, cauliflower, colour, phenolics

Procedia PDF Downloads 308

Authors: Tahir Iqbal, Mumtaz Ali, Syed Wajahat Kareem, Muhammad Harris

Abstract:

Digital information is expanding in exponential order in our life. Information that is residing online and offline are stored in huge repositories relating to every aspect of our lives. Getting the required information is a task of retrieval systems. Content based image retrieval (CBIR) is a retrieval system that retrieves the required information from repositories on the basis of the contents of the image. Time is a critical factor in retrieval system and using indexed views with CBIR system improves the time efficiency of retrieved results.

Keywords: content based image retrieval (CBIR), indexed view, color, image retrieval, cross correlation

Procedia PDF Downloads 458

Authors: Xiaoqi Jia, Qing Zhu, Hao Zhang, Su Yang

Abstract:

A gender recognition system is able to tell the gender of the given person through a few of frontal facial images. An effective gender recognition approach enables to improve the performance of many other applications, including security monitoring, human-computer interaction, image or video retrieval and so on. In this paper, we present an effective method for gender classification task in frontal facial images based on deep belief networks (DBNs), which can pre-train model and improve accuracy a little bit. Our experiments have shown that the pre-training method with DBNs for gender classification task is feasible and achieves a little improvement of accuracy on FERET and CAS-PEAL-R1 facial datasets.

Keywords: gender recognition, beep belief net-works, semi-supervised learning, greedy-layer wise RBMs

Procedia PDF Downloads 438

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 77

Authors: Qing Wu, Fei-Yan Li, Heng-Chang Zhang

Abstract:

Multitask learning support vector machines (SVMs) have recently attracted increasing research attention. Given several related tasks, the single-task learning methods trains each task separately and ignore the inner cross-relationship among tasks. However, multitask learning can capture the correlation information among tasks and achieve better performance by training all tasks simultaneously. In addition, the asymmetric squared loss function can better improve the generalization ability of the models on the most asymmetric distributed data. In this paper, we first make two assumptions on the relatedness among tasks and propose two multitask learning proximal support vector machine algorithms, named MTL-a-PSVM and EMTL-a-PSVM, respectively. MTL-a-PSVM seeks a trade-off between the maximum expectile distance for each task model and the closeness of each task model to the general model. As an extension of the MTL-a-PSVM, EMTL-a-PSVM can select appropriate kernel functions for shared information and private information. Besides, two corresponding special cases named MTL-PSVM and EMTLPSVM are proposed by analyzing the asymmetric squared loss function, which can be easily implemented by solving linear systems. Experimental analysis of three classification datasets demonstrates the effectiveness and superiority of our proposed multitask learning algorithms.

Keywords: multitask learning, asymmetric squared loss, EMTL-a-PSVM, classification

Procedia PDF Downloads 101

Authors: Thrivikraman Aswathi, S. Advaith

Abstract:

As there can be cases where the use of real data is somehow limited, such as when it is hard to get access to a large volume of real data, we need to go for synthetic data generation. This produces high-quality synthetic data while maintaining the statistical properties of a specific dataset. In the present work, a generative adversarial network (GAN) is trained to produce multivariate time series (MTS) data since the MTS is now being gathered more often in various real-world systems. Furthermore, the GAN-generated MTS data is fed into a transformer-based deep learning architecture that carries out the data categorization into predefined classes. Further, the model is evaluated across various distinct domains by generating corresponding MTS data.

Keywords: GAN, transformer, classification, multivariate time series

Procedia PDF Downloads 117

Authors: Subhadeep Mandal, Sujoy Bhattacharya, Pabitra Mitra, Diya Guha Roy, Seema Bhattacharya

Abstract:

E-commerce firms allow customers to evaluate and review the things they buy as a positive or bad experience. The e-commerce transaction processes are made up of a variety of diverse organizations and activities that operate independently but are connected together to complete the transaction (from placing an order to the goods reaching the client). After a negative shopping experience, clients frequently disregard the critical assessment of these businesses and submit their feedback on an all-over basis, which benefits certain enterprises but is tedious for others. In this article, we solely dealt with negative reviews and attempted to distinguish between negative reviews where the e-commerce firm is explicitly blamed by customers for a bad purchasing experience and other negative reviews.

Keywords: e-commerce, online shopping, customer reviews, customer behaviour, text analytics, n-grams classification

Procedia PDF Downloads 250

Authors: Phakamas Rachamontree, Theerawut Phusantisampan, Natthakorn Woravutthikul, Peerapong Pornwongthong, Malinee Sriariyanun

Abstract:

A total of 115 yeast strains isolated from local cassava processing wastes were measured for crude protein content. Among these strains, the strain MSY-2 possessed the highest protein concentration (>3.5 mg protein/mL). By using molecular identification tools, it was identified to be a strain of Pichia kudriavzevii based on similarity of D1/D2 domain of 26S rDNA region. In this study, to optimize the protein production by MSY-2 strain, Response Surface Methodology (RSM) was applied. The tested parameters were the carbon content, nitrogen content, and incubation time. Here, the value of regression coefficient (R2) = 0.7194 could be explained by the model, which is high to support the significance of the model. Under the optimal condition, the protein content was produced up to 3.77 g per L of the culture and MSY-2 strain contain 66.8 g protein per 100 g of cell dry weight. These results revealed the plausibility of applying the novel strain of yeast in single-cell protein production.

Keywords: single cell protein, response surface methodology, yeast, cassava processing waste

Procedia PDF Downloads 393

Authors: Namik Kemal Sonmez

Abstract:

Based passive remote sensing technologies have been widely used in many plant species. However, use of these techniques in orange trees is limited. In this study, the relationships between chlorophyll content (Chl) and calculated red edge (RE) and vegetation index values of the citrus leave at different growth stages were formed the basis for the analysis. Canopy reflectance by hand-held spectroradiometer and total Chl analysis at the lab were measured simultaneously, from the random samples taken from four different parts of an orange orchard. Plant materials consisted of four different age groups of 15, 20, 25, and 30 years old orange trees. Reflectance measurements were conducted between 450 and 900 nanometer (nm) wavelength at four different bands (3 visible bands and 1 near-infrared band) at the four basic physiological periods (flowering, fruit setting, fruit maturity, and dormancy) of orange trees. According to the statistical analysis conducted, there was a strong relationship between the chlorophyll content and calculated indexes (p ≤ 0.01; R²= 0.925 at red edge and R²= 0.986 at vegetation index) at the fruit setting stage of 20 years old trees. Again at this stage, fruit setting, total Chl content values among all orange trees were significantly correlated at the RE and VI with the R² values of 0.672 and 0.635 at the 0.001 level, respectively. This indicated that the relationships between Chl content and index values were very strong at this stage, in comparison to the other stages.

Keywords: spectroradiometer, citrus, chlorophyll, reflectance, index

Procedia PDF Downloads 366

Authors: Jamshid Qodirov, Sirojiddin Komolov, Ravilov Mirahmad, Olimjon Mirzayev

Abstract:

Nowadays, natural language processing tools play a crucial role in our daily lives, including various techniques with text processing. There are very advanced models in modern languages, such as English, Russian etc. But, in some languages, such as Uzbek, the NLP models have been developed recently. Thus, there are only a few NLP models in Uzbek language. Moreover, there is no such work that could show which Uzbek NLP model behaves in different situations and when to use them. This work tries to close this gap and compares the Uzbek NLP models existing as of the time this article was written. The authors try to compare the NLP models in two different scenarios: sentiment analysis and sentence similarity, which are the implementations of the two most common problems in the industry: classification and similarity. Another outcome from this work is two datasets for classification and sentence similarity in Uzbek language that we generated ourselves and can be useful in both industry and academia as well.

Keywords: NLP, benchmak, bert, vectorization

Procedia PDF Downloads 45

Authors: Chungkham Sarojnalini, Wahengbam Sarjubala Devi

Abstract:

Antioxidant properties and nutritive values of raw and cooked Pool barb, Puntius sophore (Hamilton-Buchanan) of Eastern Himalayas, India were determined. Antioxidant activity of the methanol extract of the raw, steamed, fried and curried Pool barb was evaluated by using 1,1-diphenyl-2-picrylhydrazyl (DPPH) scavenging assay. In DPPH scavenging assay the IC50 value of the raw, steamed, fried and curried Pool barb was 1.66 microgram/ml, 16.09 microgram/ml, 8.99 microgram/ml, 0.59 microgram/ml whereas the IC50 of the reference ascorbic acid was 46.66 microgram/ml. This results shows that the fish have high antioxidant activity. Protein content was found highest in raw (20.50±0.08%) and lowest in curried (18.66±0.13%). Moisture content in raw, fried and curried was 76.35±0.09, 46.27±0.14 and 57.46±0.24 respectively. Lipid content was recorded 2.46±0.14% in raw and 21.76±0.10% in curried. Ash content varies from 12.57±0.11 to 22.53±0.07%. The total aminoacids were varied from 36.79±0.02 and 288.43±0.12 mg/100 g. Eleven essential mineral elements were found abundant in all the samples. The samples had a considerable amount of Fe ranging from 152.17 to 320.39 milligram/100 gram, Ca 902.06 to 1356.02 milligram/100 gram, Zn 91.07 to 138.14 milligram/100 gram, K 193.25 to 261.56 milligram/100 gram, Mg 225.06 to 229.10 milligram/100 gram. Ni was not detected in the curried fish. The Mg and K contents were significantly decreased in frying method; however the Fe, Cu, Ca, Co and Mn content were increased significantly in all the cooked samples. The Mg and Na contents were significantly increased in curried sample and the Cr content was decreased significantly (p<0.05) in all the cooked samples.

Keywords: antioxidant property, pool barb, minerals, aminoacids, proximate composition, cooking methods

Procedia PDF Downloads 211

Authors: Dayo Samuel Banjo, Connice Trimmingham, Niloofar Yousefi, Nitin Agarwal

Abstract:

Technological advancement and its omnipresent connection have pushed humans past the boundaries and limitations of a computer screen, physical state, or geographical location. It has provided a depth of avenues that facilitate human-computer interaction that was once inconceivable such as audio and body language detection. Given the complex modularities of emotions, it becomes vital to study human-computer interaction, as it is the commencement of a thorough understanding of the emotional state of users and, in the context of social networks, the producers of multimodal information. This study first acknowledges the accuracy of classification found within multimodal emotion detection systems compared to unimodal solutions. Second, it explores the characterization of multimedia content produced based on their emotions and the coherence of emotion in different modalities by utilizing deep learning models to classify emotion across different modalities.

Keywords: affective computing, deep learning, emotion recognition, multimodal

Procedia PDF Downloads 142

Authors: Natthawut Arin

Abstract:

In forensic contexts, competency to stand trial assessments are the most common referrals. The defendants may attempt to endorse psychopathology symptoms and feign incompetent. Coaching, which can be teaching them test-taking strategies to avoid detection of psychopathological symptoms feigning. Recently, the Symptom Validity Testings (SVTs) were created to detect feigning. Moreover, the works of the literature showed that the effects of coaching on SVTs may be more robust to the effects of coaching. Thai Symptom Validity Test (SVT-Th) was designed as SVTs which demonstrated adequate psychometric properties and ability to classify between feigners and honest responders. Thus, the current study to examine the utility as the robustness of SVT-Th in the detection of feigned psychopathology. Participants consisted of 120 were recruited from undergraduate courses in psychology, randomly assigned to one of three groups. The SVT-Th was administered to those three scenario-experimental groups: (a) Uncoached group were asked to respond honestly (n=40), (b) Symptom-coached without warning group were asked to feign psychiatric symptoms to gain incompetency to stand trial (n=40), while (c) Test-coached with warning group were asked to feign psychiatric symptoms to avoid test detection but being incompetency to stand trial (n=40). Group differences were analyzed using one-way ANOVAs. The result revealed an uncoached group (M = 4.23, SD.= 5.20) had significantly lower SVT-Th mean scores than those both coached groups (M =185.00, SD.= 72.88 and M = 132.10, SD.= 54.06, respectively). Classification rates were calculated to determine the classification accuracy. Result indicated that SVT-Th had overall classification accuracy rates of 96.67% with acceptable of 95% sensitivity and 100% specificity rates. Overall, the results of the present study indicate that the SVT-Th yielded high adequate indices of accuracy and these findings suggest that the SVT-Th is robustness against coaching.

Keywords: incompetency to stand trial, coaching, robustness, classification accuracy

Procedia PDF Downloads 126