Search results for: features%20extraction

Authors: Hassan Eshkiki, Benjamin Mora

Abstract:

The Neighbor Migrating Generator is a simple and efficient approach to finding the closest potential neighbor(s) with a different label for a given instance and so without the need to calibrate any kernel settings at all. This allows determining and explaining the most important features that will influence an AI model. It can be used to either migrate a specific sample to the class decision boundary of the original model within a close neighborhood of that sample or identify global features that can help localising neighbor classes. The proposed technique works by minimizing a loss function that is divided into two components which are independently weighted according to three parameters α, β, and ω, α being self-adjusting. Results show that this approach is superior to past techniques when detecting the smallest changes in the feature space and may also point out issues in models like over-fitting.

Keywords: explainable AI, EX AI, feature importance, counterfactual explanations

Procedia PDF Downloads 120

Authors: Isa Moazen, Ali Nahvi

Abstract:

Driver drowsiness is a major cause of vehicle accidents, and facial images are highly valuable to detect drowsiness. In this paper, we perform our research via a thermal camera to record drivers' facial images on a driving simulator. A robust real-time algorithm extracts the features using horizontal and vertical integration projection, contours, contour orientations, and cropping tools. The features are included four target areas on the cheeks and forehead. Qt compiler and OpenCV are used with two cameras with different resolutions. A high-resolution thermal camera is used for fifteen subjects, and a low-resolution one is used for a person. The results are investigated by four temperature plots and evaluated by observer rating of drowsiness.

Keywords: advanced driver assistance systems, thermal imaging, driver drowsiness detection, feature extraction

Procedia PDF Downloads 99

Authors: Jawharah Alasmari

Abstract:

In verb inflection, tense marks past/present/future action, and aspect marks progressive/continues perfect/completed actions. This usage and meaning of tense and aspect differ in Arabic and English. In this research, we applied data mining methods to test the predictive function of candidate features by using our dataset of Arabic verbs in-context, and their 7 translations. Weka machine learning classifiers is used in this experiment in order to examine the key features that can be used to provide guidance to enable a translator’s appropriate English translation of the Arabic verb tense and aspect.

Keywords: Arabic verb, English translations, mining methods, Weka software

Procedia PDF Downloads 246

Authors: N. Fuad, M. N. Taib, R. Jailani, M. E. Marwan

Abstract:

This paper discusses on power spectral density (PSD) characteristics which are extracted from three-dimensional (3D) electroencephalogram (EEG) models. The EEG signal recording was conducted on 150 healthy subjects. Development of 3D EEG models involves pre-processing of raw EEG signals and construction of spectrogram images. Then, the values of maximum PSD were extracted as features from the model. These features are analysed using mean relative power (MRP) and different mean relative power (DMRP) technique to observe the pattern among different brain balancing indexes. The results showed that by implementing these techniques, the pattern of brain balancing indexes can be clearly observed. Some patterns are indicates between index 1 to index 5 for left frontal (LF) and right frontal (RF).

Keywords: power spectral density, 3D EEG model, brain balancing, mean relative power, different mean relative power

Procedia PDF Downloads 446

Authors: Olivia Rybak-Karkosz

Abstract:

This paper's purpose was to present a case study on a questioned Claude Monet's signature forensic handwriting analysis. It is an example taken from the author’s experience as a court handwriting expert. A comparative study was conducted to determine whether the signature resembles similarities (and if so, to what measure) with the features representing the writing patterns and their natural variability typical for Claude Monet. It was conducted to check whether all writing features are within the writer's normal range of variation. The paper emphasizes the difficulties and challenges encountered by the forensic handwriting expert while analysing the questioned signature.

Keywords: artist’s signatures, authenticity of an artwork, forensic handwriting analysis, graphic-comparative method

Procedia PDF Downloads 68

Authors: R. Korchiyne, A. Sbihi, S. M. Farssi, R. Touahni, M. Tahiri Alaoui

Abstract:

Trabecular bone structure is important texture in the study of osteoporosis. Legendre multifractal spectrum can reflect the complex and self-similarity characteristic of structures. The main objective of this paper is to develop a new technique of medical image classification based on Legendre multifractal spectrum. Novel features have been developed from basic geometrical properties of this spectrum in a supervised image classification. The proposed method has been successfully used to classify medical images of bone trabeculations, and could be a useful supplement to the clinical observations for osteoporosis diagnosis. A comparative study with existing data reveals that the results of this approach are concordant.

Keywords: multifractal analysis, medical image, osteoporosis, fractal dimension, Legendre spectrum, supervised classification

Procedia PDF Downloads 488

Authors: Saman Javadi, Mehdi Hashemy, Mohahammad Mahmoodi

Abstract:

Vulnerability maps are employed as an important solution in order to handle entrance of pollution into the aquifers. The common way to provide vulnerability map is DRASTIC. Meanwhile, application of the method is not easy to apply for any aquifer due to choosing appropriate constant values of weights and ranks. In this study, a new approach using k-means clustering is applied to make vulnerability maps. Four features of depth to groundwater, hydraulic conductivity, recharge value and vadose zone were considered at the same time as features of clustering. Five regions are recognized out of the case study represent zones with different level of vulnerability. The finding results show that clustering provides a realistic vulnerability map so that, Pearson’s correlation coefficients between nitrate concentrations and clustering vulnerability is obtained 61%.

Keywords: clustering, data mining, groundwater, vulnerability assessment

Procedia PDF Downloads 568

Authors: Zainab Yahaya Suleiman

Abstract:

Physical fitness, as one of the most important keys to a healthy wellbeing, is the basis of dynamic and creative intellectual activity. As a result, the fitness world is expanding every day. It is believed that a fitness centre is a place of healing and also the natural environment is vital to speedy recovery. The aim of this paper is to propose and designs a suitable location for a fitness centre in Batagarawa metropolis. Batagarawa city is enriched with four tertiary institutions with diverse commerce and culture but lacks the facility of a well-equipped fitness centre. The proposed fitness centre intends to be an organically sound centre that will make use of principles of organic architecture to create a new pleasant environment between man and his environments. Organic architecture is the science of designing a building within pleasant natural resources and features surrounding the environment. It is regarded as visual poetry and reinterpretation of nature’s principles; as well as embodies a settlement of person, place, and materials. Using organic architecture, the design was interlaced with the dynamic, organic and monumental features surrounding the environment. The city has inadequate/no facility that is considered organic where one can keep fit in a friendly, conducive and adequate location. Thus, the need for establishing a fitness centre to cater for this need cannot be over-emphasised. Conclusively, a fitness centre will be an added advantage to this fast growing centre of learning.

Keywords: organic architecture, fitness center, environment, natural resources, natural features, building design

Procedia PDF Downloads 383

Authors: Anthony D. Rhodes, Manan Goel

Abstract:

We provide a high fidelity deep learning algorithm (HyperSeg) for interactive video segmentation tasks using a dense convolutional network with context-aware skip connections and compressed, 'hypercolumn' image features combined with a convolutional tessellation procedure. In order to maintain high output fidelity, our model crucially processes and renders all image features in high resolution, without utilizing downsampling or pooling procedures. We maintain this consistent, high grade fidelity efficiently in our model chiefly through two means: (1) we use a statistically-principled, tensor decomposition procedure to modulate the number of hypercolumn features and (2) we render these features in their native resolution using a convolutional tessellation technique. For improved pixel-level segmentation results, we introduce a boundary loss function; for improved temporal coherence in video data, we include temporal image information in our model. Through experiments, we demonstrate the improved accuracy of our model against baseline models for interactive segmentation tasks using high resolution video data. We also introduce a benchmark video segmentation dataset, the VFX Segmentation Dataset, which contains over 27,046 high resolution video frames, including green screen and various composited scenes with corresponding, hand-crafted, pixel-level segmentations. Our work presents a improves state of the art segmentation fidelity with high resolution data and can be used across a broad range of application domains, including VFX pipelines and medical imaging disciplines.

Keywords: computer vision, object segmentation, interactive segmentation, model compression

Procedia PDF Downloads 96

Authors: O. I. Nkwonta, B. Dzwairo, J. Adeyemo, A. Jaiyola, N. Sawyerr, F. Otieno

Abstract:

The effective management of water resources is of great importance to ensure the supply of water resources to support changing water requirements over a selected planning horizon and in a sustainable and cost-effective way. Essentially, the purpose of the water resources planning process is to balance the available water resources in a system with the water requirements and losses to which the system is subjected. In such situations, Water resources yield and planning model can be used to solve those difficulties. It has an advantage over other models by managing model runs, developing a representative system network, modelling incremental sub-catchments, creating a variety of standard system features, special modelling features, and run result output options.

Keywords: complex, water resources, planning, cost effective and management

Procedia PDF Downloads 547

Authors: Prabhakar Sathujoda

Abstract:

Vibration analysis of a misaligned rotor coupling bearing system has been carried out while decelerating through its critical speed. The finite element method (FEM) is used to model the rotor system and simulate flexural vibrations. A flexible coupling with a frictionless joint is considered in the present work. The continuous wavelet transform is used to extract the misalignment features from the simulated time response. Subcritical speeds at one-half, one-third, and one-fourth the critical speed have appeared in the wavelet transformed vibration response of a misaligned rotor coupling bearing system. These features are also verified through a parametric study.

Keywords: Continuous Wavelet Transform, Flexible Coupling, Rotor System, Sub Critical Speed

Procedia PDF Downloads 126

Authors: M. Karthigayan, M. Rizon, Sazali Yaacob, R. Nagarajan, M. Muthukumaran, Thinaharan Ramachandran, Sargunam Thirugnanam

Abstract:

In this paper, using a set of irregular and regular ellipse fitting equations using Genetic algorithm (GA) are applied to the lip and eye features to classify the human emotions. Two South East Asian (SEA) faces are considered in this work for the emotion classification. There are six emotions and one neutral are considered as the output. Each subject shows unique characteristic of the lip and eye features for various emotions. GA is adopted to optimize irregular ellipse characteristics of the lip and eye features in each emotion. That is, the top portion of lip configuration is a part of one ellipse and the bottom of different ellipse. Two ellipse based fitness equations are proposed for the lip configuration and relevant parameters that define the emotions are listed. The GA method has achieved reasonably successful classification of emotion. In some emotions classification, optimized data values of one emotion are messed or overlapped to other emotion ranges. In order to overcome the overlapping problem between the emotion optimized values and at the same time to improve the classification, a fuzzy clustering method (FCM) of approach has been implemented to offer better classification. The GA-FCM approach offers a reasonably good classification within the ranges of clusters and it had been proven by applying to two SEA subjects and have improved the classification rate.

Keywords: ellipse fitness function, genetic algorithm, emotion recognition, fuzzy clustering

Procedia PDF Downloads 524

Authors: Carmela Cucuzzella, Jean-Pierre Chupin

Abstract:

Eco-architecture today is being assessed and judged increasingly on the basis of its environmental performance and its dedication to urgent stakes of sustainability. Architects have responded to environmental imperatives in novel ways since the 1960s. In the last two decades, however, different forms of eco-architecture practices have emerged that seem to be as dedicated to the issues of sustainability, as to their ability to 'communicate' their ecological features. The hypothesis is that some contemporary eco-architecture has been developing a characteristic 'explanatory discourse', of which it is possible to identify in buildings around the world. Some eco-architecture practices do not simply demonstrate their alignment with pressing ecological issues, rather, these buildings seem to be also driven by the urgent need to explain their ‘greenness’. The design aims specifically to teach visitors of the eco-qualities. These types of architectural practices are referred to in this paper as eco-didactic. The aim of this paper is to identify and assess this distinctive form of environmental architecture practice that aims to teach. These buildings constitute an entirely new form of design practice that places eco-messages squarely in the public realm. These eco-messages appear to have a variety of purposes: (i) to raise awareness of unsustainable quotidian habits, (ii) to become means of behavioral change, (iii) to publicly announce their responsibility through the designed eco-features, or (iv) to engage the patrons of the building into some form of sustainable interaction. To do this, a comprehensive review of Canadian eco-architecture is conducted since 1998. Their potential eco-didactic aspects are analysed through a lens of three vectors: (1) cognitive visitor experience: between the desire to inform and the poetics of form (are parts of the design dedicated to inform the visitors of the environmental aspects?); (2) formal architectural qualities: between the visibility and the invisibility of environmental features (are these eco-features clearly visible by the visitors?); and (3) communicative method for delivering eco-message: this transmission of knowledge is accomplished somewhere between consensus and dissensus as a method for disseminating the eco-message (do visitors question the eco-features or are they accepted by visitors as features that are environmental?). These architectural forms distinguish themselves in their crossing of disciplines, specifically, architecture, environmental design, and art. They also differ from other architectural practices in terms of how they aim to mobilize different publics within various urban landscapes The diversity of such buildings, from how and what they aim to communicate, to the audience they wish to engage, are all key parameters to better understand their means of knowledge transfer. Cases from the major cities across Canada are analysed, aiming to illustrate this increasing worldwide phenomenon.

Keywords: eco-architecture, public awareness, community engagement, didacticism, communication

Procedia PDF Downloads 93

Authors: Claus N. Holm, Oliver F. Grooss, Robert A. Alphinas

Abstract:

This research strives to predict the remaining product life cycle of a song on radio after it has been played for one or two months. The best results were achieved using a k-d tree to calculate the most similar songs to the test songs and use a Random Forest model to forecast radio plays. An 82.78% and 83.44% accuracy is achieved for the two time periods, respectively. This explorative research leads to over 4500 test metrics to find the best combination of models and pre-processing techniques. Other algorithms tested are KNN, MLP and CNN. The features only consist of daily radio plays and use no musical features.

Keywords: hit song science, product life cycle, machine learning, radio

Procedia PDF Downloads 126

Authors: Htay Htay Yi, Zin May Aye

Abstract:

Network security is role of the ICT environment because malicious users are continually growing that realm of education, business, and then related with ICT. The network security contravention is typically described and examined centrally based on a security event management system. The firewalls, Intrusion Detection System (IDS), and Intrusion Prevention System are becoming essential to monitor or prevent of potential violations, incidents attack, and imminent threats. In this system, the firewall rules are set only for where the system policies are needed. Dataset deployed in this system are derived from the testbed environment. The traffic as in DoS and PortScan traffics are applied in the testbed with firewall and IDS implementation. The network traffics are classified as normal or attacks in the existing testbed environment based on six machine learning classification methods applied in the system. It is required to be tested to get datasets and applied for DoS and PortScan. The dataset is based on CICIDS2017 and some features have been added. This system tested 26 features from the applied dataset. The system is to reduce false positive rates and to improve accuracy in the implemented testbed design. The system also proves good performance by selecting important features and comparing existing a dataset by machine learning classifiers.

Keywords: false negative rate, intrusion detection system, machine learning methods, performance

Procedia PDF Downloads 96

Authors: Son Nguyen, Thanh Van, Ly Le

Abstract:

Ca2+ - dependent calmodulin kinase IIa (CaMKIIa) has recently been found to associate with protein tau missorting and polymerization in Alzheimer’s Disease (AD). However, there has yet inhibitors targeting CaMKIIa to investigate the correlation between CaMKIIa activity and protein tau polymer formation. Combining virtual screening and our statistics in binding contribution scoring function (BCSF), we rationally identified potential compounds that bind to specific CaMKIIa active site and specificity-affinity distribution of the ligand within the active site. Using molecular dynamics simulation, we identified structural stability of CaMKIIa and potent inhibitors, and site-directed bonding, separating non-specific and specific molecular interaction features. Despite of variation in confirmation of simulation time, interactions of the potent inhibitors were found to be strongly associated with the unique chemical features extracted from molecular binding poses. In addition, competitive inhibitors within CaMKIIa showed an important molecular recognition pattern toward specific ligand features. Our approach combining virtual screening with BCSF may provide an universally applicable method for precise identification in the discovery of compounds.

Keywords: Alzheimer’s disease, Ca 2+ -dependent calmodulin kinase IIa, protein tau, molecular docking

Procedia PDF Downloads 247

Authors: Alua Temirbolat, Sergei Kibalnik, Zhuldyz Essimova

Abstract:

The article is devoted to research of influence of the modern literature on the consciousness of the person. Tendencies and features of the progress of the historical-cultural and artistic process at the end of XX–the beginning of XXI centuries are considered. The object of the analysis is the popular literature which has found last decades greater popularity among readers of different generations. In the article, such genres, as melodramas, female, espionage, criminal, pink, costume-historical novels, thrillers, elements, a fantasy are considered. During research, specific features of the popular literature, its difference from works of classics is revealed. On specific examples, its negative and positive influence on consciousness, psychology of the reader is shown, its role and value in a modern society are defined.

Keywords: the popular literature, the person, consciousness, a genre, psychology

Procedia PDF Downloads 269

Authors: Ben G. Weber, Michael Mateas

Abstract:

We present a case-based reasoning technique for selecting build orders in a real-time strategy game. The case retrieval process generalizes features of the game state and selects cases using domain-specific recall methods, which perform exact matching on a subset of the case features. We demonstrate the performance of the technique by implementing it as a component of the integrated agent framework of McCoy and Mateas. Our results demonstrate that the technique outperforms nearest-neighbor retrieval when imperfect information is enforced in a real-time strategy game.

Keywords: case based reasoning, real time strategy systems, requirements elicitation, requirement analyst, artificial intelligence

Procedia PDF Downloads 410

Authors: Eman A. Al-Hilo, Hawraa H. Al-Waelly

Abstract:

In this work, fractal compression (FIC) technique is introduced based on using moment features to block indexing the zero-mean range-domain blocks. The moment features have been used to speed up the IFS-matching stage. Its moments ratio descriptor is used to filter the domain blocks and keep only the blocks that are suitable to be IFS matched with tested range block. The results of tests conducted on Lena picture and Cat picture (256 pixels, resolution 24 bits/pixel) image showed a minimum encoding time (0.89 sec for Lena image and 0.78 of Cat image) with appropriate PSNR (30.01dB for Lena image and 29.8 of Cat image). The reduction in ET is about 12% for Lena and 67% for Cat image.

Keywords: fractal gray level image, fractal compression technique, iterated function system, moments feature, zero-mean range-domain block

Procedia PDF Downloads 469

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 269

Authors: Musaab Hasan, Farkhund Iqbal, Patrick C. K. Hung, Benjamin C. M. Fung, Laura Rafferty

Abstract:

In modern societies, the smart cities concept raised simultaneously with the projection towards adopting smart devices. A smart grid is an essential part of any smart city as both consumers and power utility companies benefit from the features provided by the power grid. In addition to advanced features presented by smart grids, there may also be a risk when the grids are exposed to malicious acts such as security attacks performed by terrorists. Considering advanced security measures in the design of smart meters could reduce these risks. This paper presents a security study for smart metering systems with a prototype implementation of the user interfaces for future works.

Keywords: security design, smart city, smart meter, smart grid, smart metering system

Procedia PDF Downloads 304

Authors: Sanchali Das, Swapan Debbarma

Abstract:

Music mood classification is developing in both the areas of music information retrieval (MIR) and natural language processing (NLP). Some of the Indian languages like Hindi English etc. have considerable exposure in MIR. But research in mood classification in regional language is very less. In this paper, powerful audio based feature for Kokborok Christian song is identified and mood classification task has been performed. Kokborok is an Indo-Burman language especially spoken in the northeastern part of India and also some other countries like Bangladesh, Myanmar etc. For performing audio-based classification task, useful audio features are taken out by jMIR software. There are some standard audio parameters are there for the audio-based task but as known to all that every language has its unique characteristics. So here, the most significant features which are the best fit for the database of Kokborok song is analysed. The regression-based model is used to find out the independent parameters that act as a predictor and predicts the dependencies of parameters and shows how it will impact on overall classification result. For classification WEKA 3.5 is used, and selected parameters create a classification model. And another model is developed by using all the standard audio features that are used by most of the researcher. In this experiment, the essential parameters that are responsible for effective audio based mood classification and parameters that do not significantly change for each of the Christian Kokborok songs are analysed, and a comparison is also shown between the two above model.

Keywords: Christian Kokborok song, mood classification, music information retrieval, regression

Procedia PDF Downloads 191

Authors: Punnoose A. K.

Abstract:

This paper presents an approach for identifying noisy speech recording using a multi-layer perception (MLP) trained to predict phonemes from acoustic features. Characteristics of the MLP posteriors are explored for clean speech and noisy speech at the frame level. Appropriate density functions are used to fit the softmax probability of the clean and noisy speech. A function that takes into account the ratio of the softmax probability density of noisy speech to clean speech is formulated. These phoneme independent scoring is weighted using a phoneme-specific weightage to make the scoring more robust. Simple thresholding is used to identify the noisy speech recording from the clean speech recordings. The approach is benchmarked on standard databases, with a focus on precision.

Keywords: noisy speech identification, speech pre-processing, noise robustness, feature engineering

Procedia PDF Downloads 95

Authors: Tusar Kanti Dash, Ganapati Panda

Abstract:

The evaluation of speech quality and intelligence is critical to the overall effectiveness of the Speech Enhancement Algorithms. Several intrusive and non-intrusive measures are employed to calculate these parameters. Non-Intrusive Evaluation is most challenging as, very often, the reference clean speech data is not available. In this paper, a novel non-intrusive speech evaluation measure is proposed using audio features derived from the Stockwell transform. These features are used with the Light Gradient Boosting Machine for the effective prediction of speech quality and intelligibility. The proposed model is analyzed using noisy and reverberant speech from four databases, and the results are compared with the standard Intrusive Evaluation Measures. It is observed from the comparative analysis that the proposed model is performing better than the standard Non-Intrusive models.

Keywords: non-Intrusive speech evaluation, S-transform, light GBM, speech quality, and intelligibility

Procedia PDF Downloads 228

Authors: Sanjit Kumar Paul, A. A. Mamun, M. R. Amin

Abstract:

The Linear propagation of the dust-acoustic wave in a dusty plasma consisting of Boltzmann distributed electrons and ions and mobile charge fluctuating positive dust grains has been investigated by employing the reductive perturbation method. It has been shown that the dust charge fluctuation is a source of dissipation and its responsible for the formation of the dust-acoustic waves in such a dusty plasma. The basic features of such dust-acoustic waves have been identified. It has been proposed to design a new laboratory experiment which will be able to identify the basic features of the dust-acoustic waves predicted in this theoretical investigation.

Keywords: dust acoustic waves, dusty plasma, Boltzmann distributed electrons, charge fluctuation

Procedia PDF Downloads 610

Authors: S. Sharan Sundar, Nithish B. Moudhgalya, Nidhi Bhandari, Vineeth Vijayaraghavan

Abstract:

Automatic Short Answer Grading is one of the prime applications of artificial intelligence in education. Several approaches involving the utilization of selective handcrafted features, graphical matching techniques, concept identification and mapping, complex deep frameworks, sentence embeddings, etc. have been explored over the years. However, keeping in mind the real-world application of the task, these solutions present a slight overhead in terms of computations and resources in achieving high performances. In this work, a simple and effective solution making use of elemental features based on statistical, linguistic properties, and word-based similarity measures in conjunction with tree-based classifiers and regressors is proposed. The results for classification tasks show improvements ranging from 1%-30%, while the regression task shows a stark improvement of 35%. The authors attribute these improvements to the addition of multiple similarity scores to provide ensemble of scoring criteria to the models. The authors also believe the work could reinstate that classical natural language processing techniques and simple machine learning models can be used to achieve high results for short answer grading.

Keywords: artificial intelligence, intelligent systems, natural language processing, text mining

Procedia PDF Downloads 111

Authors: Y. Xu, L. Xiong, Z. Xu

Abstract:

In order to protect data privacy, image with sensitive or private information needs to be encrypted before being outsourced to the cloud. However, this causes difficulties in image retrieval and data management. A secure image retrieval method based on orthogonal decomposition is proposed in the paper. The image is divided into two different components, for which encryption and feature extraction are executed separately. As a result, cloud server can extract features from an encrypted image directly and compare them with the features of the queried images, so that the user can thus obtain the image. Different from other methods, the proposed method has no special requirements to encryption algorithms. Experimental results prove that the proposed method can achieve better security and better retrieval precision.

Keywords: secure image retrieval, secure search, orthogonal decomposition, secure cloud computing

Procedia PDF Downloads 452

Authors: Natalia Efremova, Sergey Tarasenko

Abstract:

We propose a novel neural network architecture for visual saliency detections, which utilizes neuro physiologically plausible mechanisms for extraction of salient regions. The model has been significantly inspired by recent findings from neuro physiology and aimed to simulate the bottom-up processes of human selective attention. Two types of features were analyzed: color and direction of maximum variance. The mechanism we employ for processing those features is PCA, implemented by means of normalized Hebbian learning and the waves of spikes. To evaluate performance of our model we have conducted psychological experiment. Comparison of simulation results with those of experiment indicates good performance of our model.

Keywords: neural network models, visual saliency detection, normalized Hebbian learning, Oja's rule, psychological experiment

Procedia PDF Downloads 320

Authors: Arian Hosseini, Mahmudul Hasan

Abstract:

To address the increasing need for efficient and accurate content moderation, we propose an efficient and lightweight deep classification ensemble structure. Our approach is based on a combination of simple visual features, designed for high-accuracy classification of violent content with low false positives. Our ensemble architecture utilizes a set of lightweight models with narrowed-down color features, and we apply it to both images and videos. We evaluated our approach using a large dataset of explosion and blast contents and compared its performance to popular deep learning models such as ResNet-50. Our evaluation results demonstrate significant improvements in prediction accuracy, while benefiting from 7.64x faster inference and lower computation cost. While our approach is tailored to explosion detection, it can be applied to other similar content moderation and violence detection use cases as well. Based on our experiments, we propose a "think small, think many" philosophy in classification scenarios. We argue that transforming a single, large, monolithic deep model into a verification-based step model ensemble of multiple small, simple, and lightweight models with narrowed-down visual features can possibly lead to predictions with higher accuracy.

Keywords: deep classification, content moderation, ensemble learning, explosion detection, video processing

Procedia PDF Downloads 21

Authors: O. Nkwonta, B. Dzwairo, F. Otieno, J. Adeyemo

Abstract:

The effective management of water resources is of great importance to ensure the supply of water resources to support changing water requirements over a selected planning horizon and in a sustainable and cost-effective way. Essentially, the purpose of the water resources planning process is to balance the available water resources in a system with the water requirements and losses to which the system is subjected. In such situations, water resources yield and planning model can be used to solve those difficulties. It has an advantage over other models by managing model runs, developing a representative system network, modelling incremental sub-catchments, creating a variety of standard system features, special modelling features, and run result output options.

Keywords: complex, water resources, planning, cost effective, management

Procedia PDF Downloads 417