Abstract: Datasets or collections are becoming important assets by themselves and now they can be accepted as a primary intellectual output of a research. The quality and usage of the datasets depend mainly on the context under which they have been collected, processed, analyzed, validated, and interpreted. This paper aims to present a collection of program educational objectives mapped to student’s outcomes collected from self-study reports prepared by 32 engineering programs accredited by ABET. The manual mapping (classification) of this data is a notoriously tedious, time consuming process. In addition, it requires experts in the area, which are mostly not available. It has been shown the operational settings under which the collection has been produced. The collection has been cleansed, preprocessed, some features have been selected and preliminary exploratory data analysis has been performed so as to illustrate the properties and usefulness of the collection. At the end, the collection has been benchmarked using nine of the most widely used supervised multiclass classification techniques (Binary Relevance, Label Powerset, Classifier Chains, Pruned Sets, Random k-label sets, Ensemble of Classifier Chains, Ensemble of Pruned Sets, Multi-Label k-Nearest Neighbors and Back-Propagation Multi-Label Learning). The techniques have been compared to each other using five well-known measurements (Accuracy, Hamming Loss, Micro-F, Macro-F, and Macro-F). The Ensemble of Classifier Chains and Ensemble of Pruned Sets have achieved encouraging performance compared to other experimented multi-label classification methods. The Classifier Chains method has shown the worst performance. To recap, the benchmark has achieved promising results by utilizing preliminary exploratory data analysis performed on the collection, proposing new trends for research and providing a baseline for future studies.
Abstract: Intrusion detection systems (IDS) are the main components of network security. These systems analyze the network events for intrusion detection. The design of an IDS is through the training of normal traffic data or attack. The methods of machine learning are the best ways to design IDSs. In the method presented in this article, the pruning algorithm of C5.0 decision tree is being used to reduce the features of traffic data used and training IDS by the least square vector algorithm (LS-SVM). Then, the remaining features are arranged according to the predictor importance criterion. The least important features are eliminated in the order. The remaining features of this stage, which have created the highest level of accuracy in LS-SVM, are selected as the final features. The features obtained, compared to other similar articles which have examined the selected features in the least squared support vector machine model, are better in the accuracy, true positive rate, and false positive. The results are tested by the UNSW-NB15 dataset.
Abstract: Maternal dietary intake during lactation might affect the growth rate of an exclusive breastfed infant. The present study was conducted to evaluate the effect of maternal dietary patterns and quality during lactation on the growth of the exclusive breastfed infant. Methods: 484 healthy lactating mothers with their infant were enrolled in this study. Only exclusive breastfed infants were included in this study which was conducted in Iran. Dietary intake of lactating mothers was assessed using a validated and reliable semi-quantitative food frequency questionnaire. Diet quality indices such as alternative Healthy eating index (HEI), Dietary energy density (DED), and adherence to Mediterranean dietary pattern score, Nordic and dietary approaches to stop hypertension (DASH) eating pattern were created. Anthropometric features of infant (weight, height, and head circumference) were recorded at birth, two and four months. Results: Weight, length, weight for height and head circumference of infants at two months and four months age were mostly in the normal range among those that mothers adhered more to the HEI in lactation period (normal weight: 61%; normal height: 59%). The prevalence of stunting at four months of age among those whose mothers adhered more to the HEI was 31% lower than those with the least adherence to HEI. Mothers in the top tertiles of HEI score had the lowest frequency of having underweight infants (18% vs. 33%; P=0.03). Odds ratio of being overweight or obese at four months age was the lowest among those infants whose mothers adhered more to the HEI (OR: 0.67 vs 0.91; Ptrend=0.03). However, there was not any significant association between adherence of mothers to Mediterranean diet as well as DASH diet and Nordic eating pattern and the growth of infants (none of weight, height or head circumference). Infant weight, length, weight for height and head circumference at two months and four months did not show significant differences among different tertile categories of mothers’ DED. Conclusions: Higher diet quality indices and more adherence of lactating mother to HEI (as an indicator of diet quality) may be associated with better growth indices of the breastfed infant. However, it seems that DED of the lactating mother does not affect the growth of the breastfed infant. Adherence to the different dietary patterns such as Mediterranean, DASH or Nordic among mothers had no different effect on the growth indices of the infants. However, higher diet quality indices and more adherence of lactating mother to HEI may be associated with better growth indices of the breastfed infant. Breastfeeding is a complete way that is not affected much by the dietary patterns of the mother. However, better diet quality might be associated with better growth.
Abstract: Electricity prices have sophisticated features such as
high volatility, nonlinearity and high frequency that make forecasting
quite difficult. Electricity price has a volatile and non-random
character so that, it is possible to identify the patterns based on the
historical data. Intelligent decision-making requires accurate price
forecasting for market traders, retailers, and generation companies.
So far, many shallow-ANN (artificial neural networks) models have
been published in the literature and showed adequate forecasting
results. During the last years, neural networks with many hidden
layers, which are referred to as DNN (deep neural networks) have
been using in the machine learning community. The goal of this
study is to investigate electricity price forecasting performance of the
shallow-ANN and DNN models for the Turkish day-ahead electricity
market. The forecasting accuracy of the models has been evaluated
with publicly available data from the Turkish day-ahead electricity
market. Both shallow-ANN and DNN approach would give successful
result in forecasting problems. Historical load, price and weather
temperature data are used as the input variables for the models.
The data set includes power consumption measurements gathered
between January 2016 and December 2017 with one-hour resolution.
In this regard, forecasting studies have been carried out comparatively
with shallow-ANN and DNN models for Turkish electricity markets
in the related time period. The main contribution of this study
is the investigation of different shallow-ANN and DNN models
in the field of electricity price forecast. All models are compared
regarding their MAE (Mean Absolute Error) and MSE (Mean Square)
results. DNN models give better forecasting performance compare to
shallow-ANN. Best five MAE results for DNN models are 0.346,
0.372, 0.392, 0,402 and 0.409.
Abstract: In this paper, a planar monopole antenna for multi band applications is proposed. The antenna structure operates at three operating frequencies at 3.7, 6.2, and 13.5 GHz which cover different communication frequency ranges. The antenna consists of a quasi-modified rectangular radiating patch with a partial ground plane and two parasitic elements (open-loop-ring resonators) to serve as coupling-bridges. A stepped cut at lower corners of the radiating patch and the partial ground plane are used, to achieve the multiband features. The proposed antenna is manufactured on the FR4 substrate and is simulated and optimized using High Frequency Simulation System (HFSS). The antenna topology possesses an area of 30.5 x 30 x 1.6 mm3. The measured results demonstrate that the candidate antenna has impedance bandwidths for 10 dB return loss and operates from 3.80 – 3.90 GHz, 4.10 – 5.20 GHz, 11.2 – 11.5 GHz and from 12.5 – 14.0 GHz, which meet the requirements of the wireless local area network (WLAN), worldwide interoperability for microwave access (WiMAX), C- (Uplink) and Ku- (Uplink) band applications. Acceptable agreement is obtained between measurement and simulation results. Experimental results show that the antenna is successfully simulated and measured, and the tri-band antenna can be achieved by adjusting the lengths of the three elements and it gives good gains across all the operation bands.
Abstract: The gait pattern in people that present motor limitations foment the demand for auxiliary locomotion devices. These artifacts for movement assistance vary according to its shape, size and functional features, following the clinical applications desired. Among the ortheses of lower limbs, the ankle-foot orthesis aims to improve the ability to walk in people with different neuromuscular limitations, although they do not always answer patients' expectations for their aesthetic and functional characteristics. The purpose of this study is to explore the possibility of using new design in additive manufacturer to reproduce the shape and functional features of a ankle-foot orthesis in an efficient and modern way. Therefore, this work presents a study about the performance of the mechanical forces through the analysis of finite elements in an ankle-foot orthesis. It will be demonstrated a study of distribution of the stress on the orthopedic device in orthostatism and during the movement in the course of patient's walk.
Abstract: Flow control valves comprise a silicon flexible membrane that deflects against a substrate, usually made of glass, containing pillars, an outlet hole, and anti-stiction features. However, there is a strong interest in using silicon instead of glass as substrate material, as it would simplify the process flow by allowing the use of well controlled anisotropic etching. Moreover, specific devices demanding a bending of the substrate would also benefit from the inherent outstanding mechanical strength of monocrystalline silicon. Unfortunately, direct Si-Si bonding is not easily achieved with highly structured wafers since residual stress may prevent the good adhesion between wafers. Using a thermoplastic polymer, such as parylene, as intermediate layer is not well adapted to this design as the wafer-to-wafer alignment is critical. An alternative anodic bonding method using an intermediate borosilicate layer has been successfully tested. This layer has been deposited onto the silicon substrate. The bonding recipe has been adapted to account for the presence of the SOI buried oxide and intermediate glass layer in order not to exceed the breakdown voltage. Flow control valves dedicated to infusion of viscous fluids at very high pressure have been made and characterized. The results are compared to previous data obtained using the standard anodic bonding method.
Abstract: Speech disfluencies are common in spontaneous speech. The primary purpose of this study was to distinguish linguistic disfluencies from stuttering disfluencies in bilingual Tamil–English (TE) speaking children. The secondary purpose was to determine whether their disfluencies are mediated by native language dominance and/or on an early onset of developmental stuttering at childhood. A detailed study was carried out to identify the prosodic and acoustic features that uniquely represent the disfluent regions of speech. This paper focuses on statistical modeling of repetitions, prolongations, pauses and interjections in the speech corpus encompassing bilingual spontaneous utterances from school going children – English and Tamil. Two classifiers including Hidden Markov Models (HMM) and the Multilayer Perceptron (MLP), which is a class of feed-forward artificial neural network, were compared in the classification of disfluencies. The results of the classifiers document the patterns of disfluency in spontaneous speech samples of school-aged children to distinguish between Children Who Stutter (CWS) and Children with Language Impairment CLI). The ability of the models in classifying the disfluencies was measured in terms of F-measure, Recall, and Precision.
Abstract: Character recognition is the process of converting a text image file into editable and searchable text file. Feature Extraction is the heart of any character recognition system. The character recognition rate may be low or high depending on the extracted features. In the proposed paper, 25 features for one character are used in character recognition. Basically, there are three steps of character recognition such as character segmentation, feature extraction and classification. In segmentation step, horizontal cropping method is used for line segmentation and vertical cropping method is used for character segmentation. In the Feature extraction step, features are extracted in two ways. The first way is that the 8 features are extracted from the entire input character using eight direction chain code frequency extraction. The second way is that the input character is divided into 16 blocks. For each block, although 8 feature values are obtained through eight-direction chain code frequency extraction method, we define the sum of these 8 feature values as a feature for one block. Therefore, 16 features are extracted from that 16 blocks in the second way. We use the number of holes feature to cluster the similar characters. We can recognize the almost Myanmar common characters with various font sizes by using these features. All these 25 features are used in both training part and testing part. In the classification step, the characters are classified by matching the all features of input character with already trained features of characters.
Abstract: All currencies around the world look very different from each other. For instance, the size, color, and pattern of the paper are different. With the development of modern banking services, automatic methods for paper currency recognition become important in many applications like vending machines. One of the currency recognition architecture’s phases is Feature detection and description. There are many algorithms that are used for this phase, but they still have some disadvantages. This paper proposes a feature detection algorithm, which merges the advantages given in the current SIFT and SURF algorithms, which we call, Speeded up Robust Scale-Invariant Feature Transform (SR-SIFT) algorithm. Our proposed SR-SIFT algorithm overcomes the problems of both the SIFT and SURF algorithms. The proposed algorithm aims to speed up the SIFT feature detection algorithm and keep it robust. Simulation results demonstrate that the proposed SR-SIFT algorithm decreases the average response time, especially in small and minimum number of best key points, increases the distribution of the number of best key points on the surface of the currency. Furthermore, the proposed algorithm increases the accuracy of the true best point distribution inside the currency edge than the other two algorithms.
Abstract: Image retrieval is the most interesting technique which is being used today in our digital world. CBIR, commonly expanded as Content Based Image Retrieval is an image processing technique which identifies the relevant images and retrieves them based on the patterns that are extracted from the digital images. In this paper, two research works have been presented using CBIR. The first work provides an automated and interactive approach to the analysis of CBIR techniques. CBIR works on the principle of supervised machine learning which involves feature selection followed by training and testing phase applied on a classifier in order to perform prediction. By using feature extraction, the image transforms such as Contourlet, Ridgelet and Shearlet could be utilized to retrieve the texture features from the images. The features extracted are used to train and build a classifier using the classification algorithms such as Naïve Bayes, K-Nearest Neighbour and Multi-class Support Vector Machine. Further the testing phase involves prediction which predicts the new input image using the trained classifier and label them from one of the four classes namely 1- Normal brain, 2- Benign tumour, 3- Malignant tumour and 4- Severe tumour. The second research work includes developing a tool which is used for tumour stage identification using the best feature extraction and classifier identified from the first work. Finally, the tool will be used to predict tumour stage and provide suggestions based on the stage of tumour identified by the system. This paper presents these two approaches which is a contribution to the medical field for giving better retrieval performance and for tumour stages identification.
Abstract: This paper proposes empirical mode decomposition
(EMD) together with wavelet transform (WT) based analytic signal
for power quality (PQ) events assessment. EMD decomposes the
complex signals into several intrinsic mode functions (IMF). As
the PQ events are non stationary, instantaneous parameters have
been calculated from these IMFs using analytic signal obtained
form WT. We obtained three parameters from IMFs and then used
KNN classifier for classification of PQ disturbance. We compared
the classification of proposed method for PQ events by obtaining
the features using Hilbert transform (HT) method. The classification
efficiency using WT based analytic method is 97.5% and using HT
based analytic signal is 95.5%.
Abstract: Strategic partnerships with suppliers play a vital role for the long-term value-based supply chain. This strategic collaboration keeps still being one of the top priority of many business organizations in order to create more additional value; benefiting mainly from supplier’s specialization, capacity and innovative power, securing supply and better managing costs and quality. However, many organizations encounter difficulties in initiating, developing and managing those partnerships and many attempts result in failures. One of the reasons for such failure is the incompatibility of members of this partnership or in other words wrong supplier selection which emphasize the significance of the selection process since it is the beginning stage. An effective selection process of strategic suppliers is critical to the success of the partnership. Although there are several research studies to select the suppliers in literature, only a few of them is related to strategic supplier selection for long-term partnership. The purpose of this study is to propose a conceptual model for the selection of strategic partnership suppliers. A two-stage approach has been used in proposed model incorporating first segmentation and second selection. In the first stage; considering the fact that not all suppliers are strategically equal and instead of a long list of potential suppliers, Kraljic’s purchasing portfolio matrix can be used for segmentation. This supplier segmentation is the process of categorizing suppliers based on a defined set of criteria in order to identify types of suppliers and determine potential suppliers for strategic partnership. In the second stage, from a pool of potential suppliers defined at first phase, a comprehensive evaluation and selection can be performed to finally define strategic suppliers considering various tangible and intangible criteria. Since a long-term relationship with strategic suppliers is anticipated, criteria should consider both current and future status of the supplier. Based on an extensive literature review; strategical, operational and organizational criteria have been determined and elaborated. The result of the selection can also be used to determine suppliers who are not ready for a partnership but to be developed for strategic partnership. Since the model is based on multiple criteria for both stages, it provides a framework for further utilization of Multi-Criteria Decision Making (MCDM) techniques. The model may also be applied to a wide range of industries and involve managerial features in business organizations.
Abstract: The selection of specific landmarks for an Unmanned
Aerial Vehicles’ Visual Navigation systems based on Automatic
Landmark Recognition has significant influence on the precision of
the system’s estimated position. At the same time, manual selection
of the landmarks does not guarantee a high recognition rate, which
would also result on a poor precision. This work aims to develop an
automatic landmark selection that will take the image of the flight
area and identify the best landmarks to be recognized by the Visual
Navigation Landmark Recognition System. The criterion to select
a landmark is based on features detected by ORB or AKAZE and
edges information on each possible landmark. Results have shown
that disposition of possible landmarks is quite different from the
human perception.
Abstract: Obesity is a low-grade inflammatory disease and may lead to health problems such as hypertension, dyslipidemia, diabetes. It is also associated with important risk factors for cardiovascular diseases. This requires the detailed evaluation of obesity, particularly in children. The aim of this study is to enlighten the potential associations between lipid ratios and obesity indices and to introduce those with discriminating features among children with obesity and metabolic syndrome (MetS). A total of 408 children (aged between six and eighteen years) participated in the scope of the study. Informed consent forms were taken from the participants and their parents. Ethical Committee approval was obtained. Anthropometric measurements such as weight, height as well as waist, hip, head, neck circumferences and body fat mass were taken. Systolic and diastolic blood pressure values were recorded. Body mass index (BMI), diagnostic obesity notation model assessment index-II (D2 index), waist-to-hip, head-to-neck ratios were calculated. Total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDLChol), low-density lipoprotein cholesterol (LDLChol) analyses were performed in blood samples drawn from 110 children with normal body weight, 164 morbid obese (MO) children and 134 children with MetS. Age- and sex-adjusted BMI percentiles tabulated by World Health Organization were used to classify groups; normal body weight, MO and MetS. 15th-to-85th percentiles were used to define normal body weight children. Children, whose values were above the 99th percentile, were described as MO. MetS criteria were defined. Data were evaluated statistically by SPSS Version 20. The degree of statistical significance was accepted as p≤0.05. Mean±standard deviation values of BMI for normal body weight children, MO children and those with MetS were 15.7±1.1, 27.1±3.8 and 29.1±5.3 kg/m2, respectively. Corresponding values for the D2 index were calculated as 3.4±0.9, 14.3±4.9 and 16.4±6.7. Both BMI and D2 index were capable of discriminating the groups from one another (p≤0.01). As far as other obesity indices were considered, waist-to hip and head-to-neck ratios did not exhibit any statistically significant difference between MO and MetS groups (p≥0.05). Diagnostic obesity notation model assessment index-II was correlated with the triglycerides-to-HDL-C ratio in normal body weight and MO (r=0.413, p≤0.01 and r=0.261, (p≤0.05, respectively). Total cholesterol-to-HDL-C and LDL-C-to-HDL-C showed statistically significant differences between normal body weight and MO as well as MO and MetS (p≤0.05). The only group in which these two ratios were significantly correlated with waist-to-hip ratio was MetS group (r=0.332 and r=0.334, p≤0.01, respectively). Lack of correlation between the D2 index and the triglycerides-to-HDL-C ratio was another important finding in MetS group. In this study, parameters and ratios, whose associations were defined previously with increased cardiovascular risk or cardiac death have been evaluated along with obesity indices in children with morbid obesity and MetS. Their profiles during childhood have been investigated. Aside from the nature of the correlation between the D2 index and triglycerides-to-HDL-C ratio, total cholesterol-to-HDL-C as well as LDL-C-to- HDL-C ratios along with their correlations with waist-to-hip ratio showed that the combination of obesity-related parameters predicts better than one parameter and appears to be helpful for discriminating MO children from MetS group.
Abstract: Obesity, a disease associated with a low-grade inflammation, is a risk factor for the development of metabolic syndrome (MetS). So far, MetS risk factors such as parameters related to glucose and lipid metabolisms as well as blood pressure were considered for the evaluation of this disease. There are still some ambiguities related to the characteristic features of MetS observed particularly in pediatric population. Hormonal imbalance is also important, and quite a lot information exists about the behaviour of some hormones in adults. However, the hormonal profiles in pediatric metabolism have not been cleared yet. The aim of this study is to investigate the profiles of cortisol, insulin, and thyroid hormones in children with MetS. The study population was composed of morbid obese (MO) children without (Group 1) and with (Group 2) MetS components. WHO BMI-for age and sex percentiles were used for the classification of obesity. The values above 99 percentile were defined as morbid obesity. Components of MetS (central obesity, glucose intolerance, high blood pressure, high triacylglycerol levels, low levels of high density lipoprotein cholesterol) were determined. Anthropometric measurements were performed. Ratios as well as obesity indices were calculated. Insulin, cortisol, thyroid stimulating hormone (TSH), free T3 and free T4 analyses were performed by electrochemiluminescence immunoassay. Data were evaluated by statistical package for social sciences program. p
Abstract: From the last decades, there is a significant technological advancement in the field of robotics, and a number of modular self-reconfigurable robots were introduced that can help in space exploration, bucket to stuff, search, and rescue operation during earthquake, etc. As there are numbers of self-reconfigurable robots, choosing the optimum one is always a concern for robot user since there is an increase in available features, facilities, complexity, etc. The objective of this research work is to present a multiple attribute decision making based methodology for coding, evaluation, comparison ranking and selection of modular self-reconfigurable robots using a technique for order preferences by similarity to ideal solution approach. However, 86 attributes that affect the structure and performance are identified. A database for modular self-reconfigurable robot on the basis of different pertinent attribute is generated. This database is very useful for the user, for selecting a robot that suits their operational needs. Two visual methods namely linear graph and spider chart are proposed for ranking of modular self-reconfigurable robots. Using five robots (Atron, Smores, Polybot, M-Tran 3, Superbot), an example is illustrated, and raking of the robots is successfully done, which shows that Smores is the best robot for the operational need illustrated, and this methodology is found to be very effective and simple to use.
Abstract: Data hiding can be achieved by Steganography or invisible digital watermarking. For digital watermarking, both accurate retrieval of the embedded watermark and the integrity of the cover image are important. Medical image security in Teleradiology is one of the applications where the embedded patient record needs to be extracted with accuracy as well as the medical image integrity verified. In this research paper, the Constant Correlation Spread Spectrum digital watermarking for medical image tamper detection and accurate embedded watermark retrieval is introduced. In the proposed method, a watermark bit from a patient record is spread in a medical image sub-block such that the correlation of all watermarked sub-blocks with a spreading code, W, would have a constant value, p. The constant correlation p, spreading code, W and the size of the sub-blocks constitute the secret key. Tamper detection is achieved by flagging any sub-block whose correlation value deviates by more than a small value, ℇ, from p. The major features of our new scheme include: (1) Improving watermark detection accuracy for high-pixel depth medical images by reducing the Bit Error Rate (BER) to Zero and (2) block-level tamper detection in a single computational process with simultaneous watermark detection, thereby increasing utility with the same computational cost.
Abstract: This paper investigated the code-mixing features in Mandarin-English bilingual children in Singapore. First, it examined whether the code-mixing rate was different in Mandarin Chinese and English contexts. Second, it explored the syntactic categories of code-mixing in Singapore bilingual children. Moreover, this study investigated whether morphological information was preserved when inserting syntactic components into the matrix language. Data are derived from the Singapore Bilingual Corpus, in which the recordings and transcriptions of sixty English-Mandarin 5-to-6-year-old children were preserved for analysis. Results indicated that the rate of code-mixing was asymmetrical in the two language contexts, with the rate being significantly higher in the Mandarin context than that in the English context. The asymmetry is related to language dominance in that children are more likely to code-mix when using their nondominant language. Concerning the syntactic categories of code-mixing words in the Singaporean bilingual children, we found that noun-mixing, verb-mixing, and adjective-mixing are the three most frequently used categories in code-mixing in the Mandarin context. This pattern mirrors the syntactic categories of code-mixing in the Cantonese context in Cantonese-English bilingual children, and the general trend observed in lexical borrowing. Third, our results also indicated that English vocabularies that carry morphological information are embedded in bare forms in the Mandarin context. These findings shed light upon how bilingual children take advantage of the two languages in mixed utterances in a bilingual environment.
Abstract: The physical effects of upstream flow obstructions such
as vegetation on cross-ventilation phenomena of a building are
important for issues such as indoor thermal comfort. Modelling such
effects in Computational Fluid Dynamics simulations may also be
challenging. The aim of this work is to establish the cross-ventilation
jet behaviour in such complex terrain conditions as well as to provide
guidelines on the implementation of CFD numerical simulations in
order to model complex terrain features such as vegetation in an
efficient manner. The methodology consists of onsite measurements
on a test cell coupled with numerical simulations. It was found
that the cross-ventilation flow is highly turbulent despite the very
low velocities encountered internally within the test cells. While no
direct measurement of the jet direction was made, the measurements
indicate that flow tends to be reversed from the leeward to the
windward side. Modelling such a phenomenon proves challenging
and is strongly influenced by how vegetation is modelled. A solid
vegetation tends to predict better the direction and magnitude of the
flow than a porous vegetation approach. A simplified terrain model
was also shown to provide good comparisons with observation. The
findings have important implications on the study of cross-ventilation
in complex terrain conditions since the flow direction does not remain
trivial, as with the traditional isolated building case.