The Capacity of Mel Frequency Cepstral Coefficients for Speech Recognition
Speech recognition is of an important contribution in promoting new technologies in human computer interaction. Today, there is a growing need to employ speech technology in daily life and business activities. However, speech recognition is a challenging task that requires different stages before obtaining the desired output. Among automatic speech recognition (ASR) components is the feature extraction process, which parameterizes the speech signal to produce the corresponding feature vectors. Feature extraction process aims at approximating the linguistic content that is conveyed by the input speech signal. In speech processing field, there are several methods to extract speech features, however, Mel Frequency Cepstral Coefficients (MFCC) is the popular technique. It has been long observed that the MFCC is dominantly used in the well-known recognizers such as the Carnegie Mellon University (CMU) Sphinx and the Markov Model Toolkit (HTK). Hence, this paper focuses on the MFCC method as the standard choice to identify the different speech segments in order to obtain the language phonemes for further training and decoding steps. Due to MFCC good performance, the previous studies show that the MFCC dominates the Arabic ASR research. In this paper, we demonstrate MFCC as well as the intermediate steps that are performed to get these coefficients using the HTK toolkit.
Learning to Recommend with Negative Ratings Based on Factorization Machine
Rating prediction is an important problem for recommender systems. The task is to predict the rating for an item that a user would give. Most of the existing algorithms for the task ignore the effect of negative ratings rated by users on items, but the negative ratings have a significant impact on users’ purchasing decisions in practice. In this paper, we present a rating prediction algorithm based on factorization machines that consider the effect of negative ratings inspired by Loss Aversion theory. The aim of this paper is to develop a concave and a convex negative disgust function to evaluate the negative ratings respectively. Experiments are conducted on MovieLens dataset. The experimental results demonstrate the effectiveness of the proposed methods by comparing with other four the state-of-the-art approaches. The negative ratings showed much importance in the accuracy of ratings predictions.
K-Means Based Matching Algorithm for Multi-Resolution Feature Descriptors
Matching high dimensional features between images is computationally expensive for exhaustive search approaches in computer vision. Although the dimension of the feature can be degraded by simplifying the prior knowledge of homography, matching accuracy may degrade as a tradeoff. In this paper, we present a feature matching method based on k-means algorithm that reduces the matching cost and matches the features between images instead of using a simplified geometric assumption. Experimental results show that the proposed method outperforms the previous linear exhaustive search approaches in terms of the inlier ratio of matched pairs.
High-Fidelity 1D Dynamic Model of a Hydraulic Servo Valve Using 3D Computational Fluid Dynamics and Electromagnetic Finite Element Analysis
The dynamic performance of a 4-way solenoid operated hydraulic spool valve has been analyzed by means of a one-dimensional modeling approach capturing flow, magnetic and fluid forces, valve inertia forces, fluid compressibility, and damping. Increased model accuracy was achieved by analyzing the detailed three-dimensional electromagnetic behavior of the solenoids and flow behavior through the spool valve body for a set of relevant operating conditions, thereby allowing the accurate mapping of flow and magnetic forces on the moving valve body, in lieu of representing the respective forces by lower-order models or by means of simplistic textbook correlations. The resulting high-fidelity one-dimensional model provided the basis for specific and timely design modification eliminating experimentally observed valve oscillations.
Wavelet-Based ECG Signal Analysis and Classification
This paper presents the processing and analysis of ECG signals. The study is based on wavelet transform and uses exclusively the MATLAB environment. This study includes removing Baseline wander and further de-noising through wavelet transform and metrics such as signal-to noise ratio (SNR), Peak signal-to-noise ratio (PSNR) and the mean squared error (MSE) are used to assess the efficiency of the de-noising techniques. Feature extraction is subsequently performed whereby signal features such as heart rate, rise and fall levels are extracted and the QRS complex was detected which helped in classifying the ECG signal. The classification is the last step in the analysis of the ECG signals and it is shown that these are successfully classified as Normal rhythm or Abnormal rhythm. The final result proved the adequacy of using wavelet transform for the analysis of ECG signals.
Terrain Classification for Ground Robots Based on Acoustic Features
The motivation of our work is to detect different
terrain types traversed by a robot based on acoustic data from the
robot-terrain interaction. Different acoustic features and classifiers
were investigated, such as Mel-frequency cepstral coefficient and
Gamma-tone frequency cepstral coefficient for the feature extraction,
and Gaussian mixture model and Feed forward neural network for the
classification. We analyze the system’s performance by comparing
our proposed techniques with some other features surveyed from
distinct related works. We achieve precision and recall values between
87% and 100% per class, and an average accuracy at 95.2%. We also
study the effect of varying audio chunk size in the application phase
of the models and find only a mild impact on performance.
An Approach Based on Statistics and Multi-Resolution Representation to Classify Mammograms
One of the significant and continual public health problems in the world is breast cancer. Early detection is very important to fight the disease, and mammography has been one of the most common and reliable methods to detect the disease in the early stages. However, it is a difficult task, and computer-aided diagnosis (CAD) systems are needed to assist radiologists in providing both accurate and uniform evaluation for mass in mammograms. In this study, a multiresolution statistical method to classify mammograms as normal and abnormal in digitized mammograms is used to construct a CAD system. The mammogram images are represented by wave atom transform, and this representation is made by certain groups of coefficients, independently. The CAD system is designed by calculating some statistical features using each group of coefficients. The classification is performed by using support vector machine (SVM).
Load Forecasting in Microgrid Systems with R and Cortana Intelligence Suite
Energy production optimization has been traditionally very important for utilities in order to improve resource consumption. However, load forecasting is a challenging task, as there are a large number of relevant variables that must be considered, and several strategies have been used to deal with this complex problem. This is especially true also in microgrids where many elements have to adjust their performance depending on the future generation and consumption conditions. The goal of this paper is to present a solution for short-term load forecasting in microgrids, based on three machine learning experiments developed in R and web services built and deployed with different components of Cortana Intelligence Suite: Azure Machine Learning, a fully managed cloud service that enables to easily build, deploy, and share predictive analytics solutions; SQL database, a Microsoft database service for app developers; and PowerBI, a suite of business analytics tools to analyze data and share insights. Our results show that Boosted Decision Tree and Fast Forest Quantile regression methods can be very useful to predict hourly short-term consumption in microgrids; moreover, we found that for these types of forecasting models, weather data (temperature, wind, humidity and dew point) can play a crucial role in improving the accuracy of the forecasting solution. Data cleaning and feature engineering methods performed in R and different types of machine learning algorithms (Boosted Decision Tree, Fast Forest Quantile and ARIMA) will be presented, and results and performance metrics discussed.
Evaluation Framework for Investments in Rail Infrastructure Projects
Transport infrastructures are high-cost, long-term investments that serve as vital foundations for the operation of a region or nation and are essential to a country’s or business’s economic development and prosperity, by improving well-being and generating jobs and income. The development of appropriate financing options is of key importance in the decision making process in order develop viable transport infrastructures. The development of transport infrastructure has increasingly been shifting toward alternative methods of project financing such as Public Private Partnership (PPPs) and hybrid forms. In this paper, a methodological decision-making framework based on the evaluation of the financial viability of transportation infrastructure for different financial schemes is presented. The framework leads to an assessment of the financial viability which can be achieved by performing various financing scenarios analyses. To illustrate the application of the proposed methodology, a case study of rail transport infrastructure financing scenario analysis in Greece is developed.
Forensic Speaker Verification in Noisy Environmental by Enhancing the Speech Signal Using ICA Approach
We propose a system to real environmental noise and
channel mismatch for forensic speaker verification systems. This
method is based on suppressing various types of real environmental
noise by using independent component analysis (ICA) algorithm.
The enhanced speech signal is applied to mel frequency cepstral
coefficients (MFCC) or MFCC feature warping to extract the
essential characteristics of the speech signal. Channel effects are
reduced using an intermediate vector (i-vector) and probabilistic
linear discriminant analysis (PLDA) approach for classification. The
proposed algorithm is evaluated by using an Australian forensic voice
comparison database, combined with car, street and home noises
from QUT-NOISE at a signal to noise ratio (SNR) ranging from -10
dB to 10 dB. Experimental results indicate that the MFCC feature
warping-ICA achieves a reduction in equal error rate about (48.22%,
44.66%, and 50.07%) over using MFCC feature warping when the
test speech signals are corrupted with random sessions of street, car,
and home noises at -10 dB SNR.
Towards a Complete Automation Feature Recognition System for Sheet Metal Manufacturing
Sheet metal processing is automated, but the step from product models to the production machine control still requires human intervention. This may cause time consuming bottlenecks in the production process and increase the risk of human errors. In this paper we present a system, which automatically recognizes features from the CAD-model of the sheet metal product. By using these features, the system produces a complete model of the particular sheet metal product. Then the model is used as an input for the sheet metal processing machine. Currently the system is implemented, capable to recognize more than 11 of the most common sheet metal structural features, and the procedure is fully automated. This provides remarkable savings in the production time, and protects against the human errors. This paper presents the developed system architecture, applied algorithms and system software implementation and testing.
Feature Selection and Predictive Modeling of Housing Data Using Random Forest
Predictive data analysis and modeling involving machine learning techniques become challenging in presence of too many explanatory variables or features. Presence of too many features in machine learning is known to not only cause algorithms to slow down, but they can also lead to decrease in model prediction accuracy. This study involves housing dataset with 79 quantitative and qualitative features that describe various aspects people consider while buying a new house. Boruta algorithm that supports feature selection using a wrapper approach build around random forest is used in this study. This feature selection process leads to 49 confirmed features which are then used for developing predictive random forest models. The study also explores five different data partitioning ratios and their impact on model accuracy are captured using coefficient of determination (r-square) and root mean square error (rsme).
Data Quality Enhancement with String Length Distribution
Recently, collectable manufacturing data are rapidly
increasing. On the other hand, mega recall is getting serious as
a social problem. Under such circumstances, there are increasing
needs for preventing mega recalls by defect analysis such as
root cause analysis and abnormal detection utilizing manufacturing
data. However, the time to classify strings in manufacturing data
by traditional method is too long to meet requirement of quick
defect analysis. Therefore, we present String Length Distribution
Classification method (SLDC) to correctly classify strings in a short
time. This method learns character features, especially string length
distribution from Product ID, Machine ID in BOM and asset list.
By applying the proposal to strings in actual manufacturing data, we
verified that the classification time of strings can be reduced by 80%.
As a result, it can be estimated that the requirement of quick defect
analysis can be fulfilled.
The Modulation of Self-interest Instruction on the Fair-Proposing Behavior in Ultimatum Game
Ultimatum game is an experimental paradigm to study human decision making. There are two players, a proposer and a responder, to split a fixed amount of money. According to the traditional economic theory on ultimatum game, proposer should propose the selfish offers to responder as much as possible to maximize proposer’s own outcomes. However, most evidences had showed that people chose more fair offers, hence two hypotheses – fairness favoring and strategic concern were proposed. In current study, we induced the motivation in participants to be either selfish or altruistic, and manipulated the task variables, the stake sizes (NT$100, 1000, 10000) and the share sizes (the 40%, 30%, 20%, 10% of the sum as selfish offers, and the 60%, 70%, 80%, 90% of the sum as altruistic offers), to examine the two hypotheses. The results showed that most proposers chose more fair offers with longer reaction times (RTs) no matter in choosing between the fair and selfish offers, or between the fair and altruistic offers. However, the proposers received explicit self-interest instruction chose more selfish offers accompanied with longer RTs in choosing between the fair and selfish offers. Therefore, the results supported the strategic concern hypothesis that previous proposers choosing the fair offers might be resulted from the fear of rejection by responders. Proposers would become more self-interest if the fear of being rejected is eliminated.
sEMG Interface Design for Locomotion Identification
Surface electromyographic (sEMG) signal has the potential to identify the human activities and intention. This potential is further exploited to control the artificial limbs using the sEMG signal from residual limbs of amputees. The paper deals with the development of multichannel cost efficient sEMG signal interface for research application, along with evaluation of proposed class dependent statistical approach of the feature selection method. The sEMG signal acquisition interface was developed using ADS1298 of Texas Instruments, which is a front-end interface integrated circuit for ECG application. Further, the sEMG signal is recorded from two lower limb muscles for three locomotions namely: Plane Walk (PW), Stair Ascending (SA), Stair Descending (SD). A class dependent statistical approach is proposed for feature selection and also its performance is compared with 12 preexisting feature vectors. To make the study more extensive, performance of five different types of classifiers are compared. The outcome of the current piece of work proves the suitability of the proposed feature selection algorithm for locomotion recognition, as compared to other existing feature vectors. The SVM Classifier is found as the outperformed classifier among compared classifiers with an average recognition accuracy of 97.40%. Feature vector selection emerges as the most dominant factor affecting the classification performance as it holds 51.51% of the total variance in classification accuracy. The results demonstrate the potentials of the developed sEMG signal acquisition interface along with the proposed feature selection algorithm.
An Adaptive Dimensionality Reduction Approach for Hyperspectral Imagery Semantic Interpretation
With the development of HyperSpectral Imagery
(HSI) technology, the spectral resolution of HSI became denser,
which resulted in large number of spectral bands, high correlation
between neighboring, and high data redundancy. However, the
semantic interpretation is a challenging task for HSI analysis
due to the high dimensionality and the high correlation of the
different spectral bands. In fact, this work presents a dimensionality
reduction approach that allows to overcome the different issues
improving the semantic interpretation of HSI. Therefore, in order
to preserve the spatial information, the Tensor Locality Preserving
Projection (TLPP) has been applied to transform the original HSI.
In the second step, knowledge has been extracted based on the
adjacency graph to describe the different pixels. Based on the
transformation matrix using TLPP, a weighted matrix has been
constructed to rank the different spectral bands based on their
contribution score. Thus, the relevant bands have been adaptively
selected based on the weighted matrix. The performance of the
presented approach has been validated by implementing several
experiments, and the obtained results demonstrate the efficiency
of this approach compared to various existing dimensionality
reduction techniques. Also, according to the experimental results,
we can conclude that this approach can adaptively select the
relevant spectral improving the semantic interpretation of HSI.
Diagnosis of Diabetes Using Computer Methods: Soft Computing Methods for Diabetes Detection Using Iris
Complementary and Alternative Medicine (CAM) techniques are quite popular and effective for chronic diseases. Iridology is more than 150 years old CAM technique which analyzes the patterns, tissue weakness, color, shape, structure, etc. for disease diagnosis. The objective of this paper is to validate the use of iridology for the diagnosis of the diabetes. The suggested model was applied in a systemic disease with ocular effects. 200 subject data of 100 each diabetic and non-diabetic were evaluated. Complete procedure was kept very simple and free from the involvement of any iridologist. From the normalized iris, the region of interest was cropped. All 63 features were extracted using statistical, texture analysis, and two-dimensional discrete wavelet transformation. A comparison of accuracies of six different classifiers has been presented. The result shows 89.66% accuracy by the random forest classifier.
Reduction of False Positives in Head-Shoulder Detection Based on Multi-Part Color Segmentation
The paper presents a method that utilizes figure-ground color segmentation to extract effective global feature in terms of false positive reduction in the head-shoulder detection. Conventional detectors that rely on local features such as HOG due to real-time operation suffer from false positives. Color cue in an input image provides salient information on a global characteristic which is necessary to alleviate the false positives of the local feature based detectors. An effective approach that uses figure-ground color segmentation has been presented in an effort to reduce the false positives in object detection. In this paper, an extended version of the approach is presented that adopts separate multipart foregrounds instead of a single prior foreground and performs the figure-ground color segmentation with each of the foregrounds. The multipart foregrounds include the parts of the head-shoulder shape and additional auxiliary foregrounds being optimized by a search algorithm. A classifier is constructed with the feature that consists of a set of the multiple resulting segmentations. Experimental results show that the presented method can discriminate more false positive than the single prior shape-based classifier as well as detectors with the local features. The improvement is possible because the presented approach can reduce the false positives that have the same colors in the head and shoulder foregrounds.
Facial Recognition on the Basis of Facial Fragments
There are many articles that attempt to establish the role of different facial fragments in face recognition. Various approaches are used to estimate this role. Frequently, authors calculate the entropy corresponding to the fragment. This approach can only give approximate estimation. In this paper, we propose to use a more direct measure of the importance of different fragments for face recognition. We propose to select a recognition method and a face database and experimentally investigate the recognition rate using different fragments of faces. We present two such experiments in the paper. We selected the PCNC neural classifier as a method for face recognition and parts of the LFW (Labeled Faces in the Wild) face database as training and testing sets. The recognition rate of the best experiment is comparable with the recognition rate obtained using the whole face.
Perception of Neighbourhood-Level Built Environment in Relation to Youth Physical Activity in Malaysia
Neighbourhood environment walkability on reported physical activity (PA) levels of students of Universiti Sains Malaysia (USM) in Malaysia. Compared with previous generations, today’s young people spend less time playing outdoors and have lower participation rates in PA. Research suggests that negative perceptions of neighbourhood walkability may be a potential barrier to adolescents’ PA. The sample consisted of 200 USM students (to 24 years old) who live outside of the main campus and engage in PA in sport halls and sport fields of USM. The data were analysed using the t-test, binary logistic regression, and discriminant analysis techniques. The present study found that youth PA was affected by neighbourhood environment walkability factors, including neighbourhood infrastructures, neighbourhood safety (crime), and recreation facilities, as well as street characteristics and neighbourhood design variables such as facades of sidewalks, roadside trees, green spaces, and aesthetics. The finding also illustrated that active students were influenced by street connectivity, neighbourhood infrastructures, recreation facilities, facades of sidewalks, and aesthetics, whereas students in the less active group were affected by access to destinations, neighbourhood safety (crime), and roadside trees and green spaces for their PAs. These results report which factors of built environments have more effect on youth PA and they message to the public to create more awareness about the benefits of PA on youth health.
Feature Extractions of EMG Signals during a Constant Workload Pedaling Exercise
Electromyography (EMG) is one of the important indicators during exercise, as it is closely related to the level of muscle activations. This work quantifies the muscle conditions of the lower limbs in a constant workload exercise. Surface EMG signals of the vastus laterals (VL), vastus medialis (VM), rectus femoris (RF), gastrocnemius medianus (GM), gastrocnemius lateral (GL) and Soleus (SOL) were recorded from fourteen healthy males. The EMG signals were segmented in two phases: activation segment (AS) and relaxation segment (RS). Period entropy (PE), peak count (PC), zero crossing (ZC), wave length (WL), mean power frequency (MPF), median frequency (MDF) and root mean square (RMS) are calculated to provide the quantitative information of the measured EMG segments. The outcomes reveal that the PE, PC, ZC and RMS have significantly changed (p<.001); WL presents moderately changed (p<.01); MPF and MDF show no changed (p>.05) during exercise. The results also suggest that the RS is also preferred for performance evaluation, while the results of the extracted features in AS are usually affected directly by the amplitudes. It is further found that the VL exhibits the most significant changes within six muscles during pedaling exercise. The proposed work could be applied to quantify the stamina analysis and to predict the instant muscle status in athletes.
Finite Element Analysis of Ball-Joint Boots under Environmental and Endurance Tests
Ball joints support and guide certain automotive parts that move relative to the frame of the vehicle. Such ball joints are covered and protected from dust, mud, and other interfering materials by ball-joint boots made of rubber—a flexible and near-incompressible material. The boots may experience twisting and bending deformations because of the motion of the joint arm. Thus, environmental and endurance tests of ball-joint boots apply both bending and twisting deformations. In this study, environmental and endurance testing was simulated via the finite element method performed by using a commercial software package. The ranges of principal stress and principal strain values that are known to directly affect the fatigue lives of the parts were sought. By defining these ranges, the number of iterative tests and modifications of the materials and dimensions of the boot can be decreased. Therefore, instead of performing actual part tests, manufacturers can perform standard fatigue tests in trials of different materials by applying only the defined range of stress or strain values.
Learners’ Violent Behaviour and Drug Abuse as Major Causes of Tobephobia in Schools
Many schools throughout the world are facing constant pressure to cope with the violence and drug abuse of learners who show little or no respect for acceptable and desirable social norms. These delinquent learners tend to harbour feelings of being beyond reproach because they strongly believe that it is well within their rights to engage in violent and destructive behaviour. Knives, guns, and other weapons appear to be more readily used by them on the school premises than before. It is known that learners smoke, drink alcohol, and use drugs during school hours, hence, their ability to concentrate, work, and learn, is affected. They become violent and display disruptive behaviour in their classrooms as well as on the school premises, and this atrocious behaviour makes it possible for drug dealers and gangsters to gain access onto the school premises. The primary purpose of this exploratory quantitative study was therefore to establish how tobephobia (TBP), caused by school violence and drug abuse, affects teaching and learning in schools. The findings of this study affirmed that poor discipline resulted in producing poor quality education. Most of the teachers in this study agreed that educating learners who consumed alcohol and other drugs on the school premises resulted in them suffering from TBP. These learners are frequently abusive and disrespectful, and resort to violence to seek attention. As a result, teachers feel extremely demotivated and suffer from high levels of anxiety and stress. The word TBP will surely be regarded as a blessing by many teachers throughout the world because finally, there is a word that will make people sit up and listen to their problems that cause real fear and anxiety in schools.
Product Feature Modelling for Integrating Product Design and Assembly Process Planning
This paper describes a part of the integrating work between assembly design and assembly process planning domains (APP). The work is based, in its first stage, on modelling assembly features to support APP. A multi-layer architecture, based on feature-based modelling, is proposed to establish a dynamic and adaptable link between product design using CAD tools and APP. The proposed approach is based on deriving “specific function” features from the “generic” assembly and form features extracted from the CAD tools. A hierarchal structure from “generic” to “specific” and from “high level geometrical entities” to “low level geometrical entities” is proposed in order to integrate geometrical and assembly data extracted from geometrical and assembly modelers to the required processes and resources in APP. The feature concept, feature-based modelling, and feature recognition techniques are reviewed.
The Growth of E-Commerce and Online Dispute Resolution in Developing Nations: An Analysis
Online dispute resolution has been identified in many countries as a viable alternative for resolving conflicts which have arisen in the so-called digital age. This system of dispute resolution is developing alongside the Internet, and as new types of transactions are made possible by our increased connectivity, new ways of resolving disputes must be explored. Developed nations, such as the United States of America and the European Union, have been involved in creating these online dispute resolution mechanisms from the outset, and currently have sophisticated systems in place to deal with conflicts arising in a number of different fields, such as e-commerce, domain name disputes, labour disputes and conflicts arising from family law. Specifically, in the field of e-commerce, the Internet’s borderless nature has served as a way to promote cross-border trade, and has created a global marketplace. Participation in this marketplace boosts a country’s economy, as new markets are now available, and consumers can transact from anywhere in the world. It would be especially advantageous for developing nations to be a part of this global marketplace, as it could stimulate much-needed investment in these nations, and encourage international co-operation and trade. However, for these types of transactions to proliferate, an effective system for resolving the inevitable disputes arising from such an increase in e-commerce is needed. Online dispute resolution scholarship and practice is flourishing in developed nations, and it is clear that the gap is widening between developed and developing nations in this regard. The potential for implementing online dispute resolution in developing countries has been discussed, but there are a number of obstacles that have thus far prevented its continued development. This paper aims to evaluate the various political, infrastructural and socio-economic challenges faced in developing nations, and to question how these have impacted the acceptance and development of online dispute resolution, scholarship and training of online dispute resolution practitioners and, ultimately, developing nations’ readiness to participate in cross-border e-commerce.
Image Retrieval Based on Multi-Feature Fusion for Heterogeneous Image Databases
Selecting an appropriate image representation is the most important factor in implementing an effective Content-Based Image Retrieval (CBIR) system. This paper presents a multi-feature fusion approach for efficient CBIR, based on the distance distribution of features and relative feature weights at the time of query processing. It is a simple yet effective approach, which is free from the effect of features' dimensions, ranges, internal feature normalization and the distance measure. This approach can easily be adopted in any feature combination to improve retrieval quality. The proposed approach is empirically evaluated using two benchmark datasets for image classification (a subset of the Corel dataset and Oliva and Torralba) and compared with existing approaches. The performance of the proposed approach is confirmed with the significantly improved performance in comparison with the independently evaluated baseline of the previously proposed feature fusion approaches.
A Spatial Hypergraph Based Semi-Supervised Band Selection Method for Hyperspectral Imagery Semantic Interpretation
Hyperspectral imagery (HSI) typically provides a
wealth of information captured in a wide range of the
electromagnetic spectrum for each pixel in the image. Hence, a
pixel in HSI is a high-dimensional vector of intensities with a
large spectral range and a high spectral resolution. Therefore, the
semantic interpretation is a challenging task of HSI analysis. We
focused in this paper on object classification as HSI semantic
interpretation. However, HSI classification still faces some issues,
among which are the following: The spatial variability of spectral
signatures, the high number of spectral bands, and the high cost
of true sample labeling. Therefore, the high number of spectral
bands and the low number of training samples pose the problem of
the curse of dimensionality. In order to resolve this problem, we
propose to introduce the process of dimensionality reduction trying
to improve the classification of HSI. The presented approach is a
semi-supervised band selection method based on spatial hypergraph
embedding model to represent higher order relationships with
different weights of the spatial neighbors corresponding to the
centroid of pixel. This semi-supervised band selection has been
developed to select useful bands for object classification. The
presented approach is evaluated on AVIRIS and ROSIS HSIs
and compared to other dimensionality reduction methods. The
experimental results demonstrate the efficacy of our approach
compared to many existing dimensionality reduction methods for
Fusion of Finger Inner Knuckle Print and Hand Geometry Features to Enhance the Performance of Biometric Verification System
With the advent of modern computing technology, there is an increased demand for developing recognition systems that have the capability of verifying the identity of individuals. Recognition systems are required by several civilian and commercial applications for providing access to secured resources. Traditional recognition systems which are based on physical identities are not sufficiently reliable to satisfy the security requirements due to the use of several advances of forgery and identity impersonation methods. Recognizing individuals based on his/her unique physiological characteristics known as biometric traits is a reliable technique, since these traits are not transferable and they cannot be stolen or lost. Since the performance of biometric based recognition system depends on the particular trait that is utilized, the present work proposes a fusion approach which combines Inner knuckle print (IKP) trait of the middle, ring and index fingers with the geometrical features of hand. The hand image captured from a digital camera is preprocessed to find finger IKP as region of interest (ROI) and hand geometry features. Geometrical features are represented as the distances between different key points and IKP features are extracted by applying local binary pattern descriptor on the IKP ROI. The decision level AND fusion was adopted, which has shown improvement in performance of the combined scheme. The proposed approach is tested on the database collected at our institute. Proposed approach is of significance since both hand geometry and IKP features can be extracted from the palm region of the hand. The fusion of these features yields a false acceptance rate of 0.75%, false rejection rate of 0.86% for verification tests conducted, which is less when compared to the results obtained using individual traits. The results obtained confirm the usefulness of proposed approach and suitability of the selected features for developing biometric based recognition system based on features from palmar region of hand.
A Hybrid Multi-Criteria Hotel Recommender System Using Explicit and Implicit Feedbacks
Recommender systems, also known as recommender engines, have become an important research area and are now being applied in various fields. In addition, the techniques behind the recommender systems have been improved over the time. In general, such systems help users to find their required products or services (e.g. books, music) through analyzing and aggregating other users’ activities and behavior, mainly in form of reviews, and making the best recommendations. The recommendations can facilitate user’s decision making process. Despite the wide literature on the topic, using multiple data sources of different types as the input has not been widely studied. Recommender systems can benefit from the high availability of digital data to collect the input data of different types which implicitly or explicitly help the system to improve its accuracy. Moreover, most of the existing research in this area is based on single rating measures in which a single rating is used to link users to items. This paper proposes a highly accurate hotel recommender system, implemented in various layers. Using multi-aspect rating system and benefitting from large-scale data of different types, the recommender system suggests hotels that are personalized and tailored for the given user. The system employs natural language processing and topic modelling techniques to assess the sentiment of the users’ reviews and extract implicit features. The entire recommender engine contains multiple sub-systems, namely users clustering, matrix factorization module, and hybrid recommender system. Each sub-system contributes to the final composite set of recommendations through covering a specific aspect of the problem. The accuracy of the proposed recommender system has been tested intensively where the results confirm the high performance of the system.
A Feasibility and Implementation Model of Small-Scale Hydropower Development for Rural Electrification in South Africa: Design Chart Development
Small scale hydropower used to play a very important role in the provision of energy to urban and rural areas of South Africa. The national electricity grid, however, expanded and offered cheap, coal generated electricity and a large number of hydropower systems were decommissioned. Unfortunately, large numbers of households and communities will not be connected to the national electricity grid for the foreseeable future due to high cost of transmission and distribution systems to remote communities due to the relatively low electricity demand within rural communities and the allocation of current expenditure on upgrading and constructing of new coal fired power stations. This necessitates the development of feasible alternative power generation technologies. A feasibility and implementation model was developed to assist in designing and financially evaluating small-scale hydropower (SSHP) plants. Several sites were identified using the model. The SSHP plants were designed for the selected sites and the designs for the different selected sites were priced using pricing models (civil, mechanical and electrical aspects). Following feasibility studies done on the designed and priced SSHP plants, a feasibility analysis was done and a design chart developed for future similar potential SSHP plant projects. The methodology followed in conducting the feasibility analysis for other potential sites consisted of developing cost and income/saving formulae, developing net present value (NPV) formulae, Capital Cost Comparison Ratio (CCCR) and levelised cost formulae for SSHP projects for the different types of plant installations. It included setting up a model for the development of a design chart for a SSHP, calculating the NPV, CCCR and levelised cost for the different scenarios within the model by varying different parameters within the developed formulae, setting up the design chart for the different scenarios within the model and analyzing and interpreting results. From the interpretation of the develop design charts for feasible SSHP in can be seen that turbine and distribution line cost are the major influences on the cost and feasibility of SSHP. High head, short transmission line and islanded mini-grid SSHP installations are the most feasible and that the levelised cost of SSHP is high for low power generation sites. The main conclusion from the study is that the levelised cost of SSHP projects indicate that the cost of SSHP for low energy generation is high compared to the levelised cost of grid connected electricity supply; however, the remoteness of SSHP for rural electrification and the cost of infrastructure to connect remote rural communities to the local or national electricity grid provides a low CCCR and renders SSHP for rural electrification feasible on this basis.