Search

Knowledge graphs (KGs) inherently lack reasoning ability which limits their effectiveness for tasks such as question-answering and query expansion. KG embedding (KGE) is a predominant approach where proximity between relations and entities in the embedding space is used for reasoning over KGs. Most existing KGE approaches use structural information of triplets and disregard contextual information which could be crucial to learning long-term relations between entities. Moreover, KGE approaches mostly use discriminative models which require both positive and negative samples to learn a decision boundary. KGs, by contrast, contain only positive samples, necessitating that negative samples are generated by replacing the head/tail of predicates with randomly-chosen entities. They are thus usually irrational and easily discriminable from positive samples, which can prevent the learning of sufficiently robust classifiers. To address the shortcomings, we propose to learn contextualized KGE using pretrained adversarial networks. We assume multi-hop relational paths(mh-RPs) as textual sequences for competitively learning discriminator-based KGE against the negative mh-RP generator. We use a pre-trained ELECTRA model and feed it with relational paths. We employ a generator to corrupt randomly-chosen entities with plausible alternatives and a discriminator to predict whether an entity is corrupted or not. We perform experiments on multiple benchmark knowledge graphs and the results show that our proposed KG-ELECTRA model outperforms BERT in link prediction.

Visual attribution (VA) in relation to medical images is an essential aspect of modern automation-assisted diagnosis. Since it is generally not straightforward to obtain pixel-level ground-truth labelling of medical images, classification-based interpretation approaches have become the de facto standard for automated diagnosis, in which the ability of classifiers to make categorical predictions based on class-salient regions is harnessed within the learning algorithm. Such regions, however, typically constitute only a small subset of the full range of features of potential medical interest. They may hence not be useful for VA of medical images where capturing all of the disease evidence is a critical requirement. This hence motivates the proposal of a novel strategy for visual attribution that is not reliant on image classification. We instead obtain normal counterparts of abnormal images and find discrepancy maps between the two. To perform the abnormal-to-normal mapping in unsupervised way, we employ a Cycle-Consistency Generative Adversarial Network, thereby formulating visual attribution in terms of a discrepancy map that, when subtracted from the abnormal image, makes it indistinguishable from the counterpart normal image. Experiments are performed on three datasets including a synthetic, Alzheimer’s disease Neuro imaging Initiative and, BraTS dataset. We outperform baseline and related methods in both experiments.

Artificial intelligence systems are becoming ubiquitous in everyday life as well as in high-risk environments, such as autonomous driving, medical treatment, and medicine. The opaque nature of the deep neural network raises concerns about its adoption in high-risk environments. It is important for researchers to explain how these models reach their decisions. Most of the existing methods rely on softmax to explain model decisions. However, softmax is shown to be often misleading, particularly giving unjustified high confidence even for samples far from the training data. To overcome this shortcoming, we propose Bayesian model uncertainty for producing counterfactual explanations. In this paper, we compare the counterfactual explanation of models based on Bayesian uncertainty and softmax score. This work predictively produces minimal important features, which maximally change classifier output to explain the decision-making process of the Bayesian model. We used MNIST and Caltech Bird 2011 datasets for experiments. The results show that the Bayesian model outperforms the softmax model and produces more concise and human-understandable counterfactuals.

Screening via low-dose Computer Tomography (CT) has been shown to reduce lung cancer mortality rates by at least 20%. However, the assessment of large numbers of CT scans by radiologists is cost intensive, and potentially produces varying and inconsistent results for differing radiologists (and also for temporally-separated assessments by the same radiologist). To overcome these challenges, computer aided diagnosis systems based on deep learning methods have proved an effective in automatic detection and classification of lung cancer.\ud \ud Latterly, interest has focused on the full utilization of the 3D information in CT scans using 3D-CNNs and related approaches. However, such approaches do not intrinsically correlate size and shape information between slices. In this work, an innovative approach to Multi-view Convolutional Recurrent Neural Networks (MV-CRecNet) is proposed that exploits shape, size and cross-slice variations while learning to identify lung cancer nodules from CT scans. The multiple-views that are passed to the model ensure better generalization and the learning of robust features.\ud \ud We evaluate the proposed MV-CRecNet model on the reference Lung Image Database Consortium and Image Database Resource Initiative and Early Lung Cancer Action Program datasets; six evaluation metrics are applied to eleven comparison models for testing. Results demonstrate that proposed methodology outperforms all of the models against all of the evaluation metrics.

This article presents an efficient fingerprint identification system that implements an initial classification for search-space reduction followed by minutiae neighbor-based feature encoding and matching. The current state-of-the-art fingerprint classification methods use a deep convolutional neural network (DCNN) to assign confidence for the classification prediction, and based on this prediction, the input fingerprint is matched with only the subset of the database that belongs to the predicted class. It can be observed for the DCNNs that as the architectures deepen, the farthest layers of the network learn more abstract information from the input images that result in higher prediction accuracies. However, the downside is that the DCNNs are data hungry and require lots of annotated (labeled) data to learn generalized network parameters for deeper layers. In this article, a shallow multifeature view CNN (SMV-CNN) fingerprint classifier is proposed that extracts: 1) fine-grained features from the input image and 2) abstract features from explicitly derived representations obtained from the input image. The multifeature views are fed to a fully connected neural network (NN) to compute a global classification prediction. The classification results show that the SMV-CNN demonstrated an improvement of 2.8% when compared to baseline CNN consisting of a single grayscale view on an open-source database. Moreover, in comparison with the state-of-the-art residual network (ResNet-50) image classification model, the proposed method performs comparably while being less complex and more efficient during training. The result of classification-based fingerprint identification has shown that the search space is reduced by over 50% without degradation of identification accuracies.

With the increase in the culture of social media and netizen, every day, millions of comments are posted on the uploaded posts. The use of abusive language in user comments has been increased rapidly. Abusive language in online comments initiates cyber-bullying that targets individuals (celebrity, politician, and product) and a group of people (specific country, age, and religion). It is important to detect and analyze abusive language from online comments automatically. There have been several attempts in the literature to detect abusive language for English. In this study, we perform abusive language detection from Urdu and Roman Urdu comments using five diverse ML models (NB, SVM, IBK, Logistic, and JRip) and four DL models (CNN, LSTM, BLSTM, and CLSTM). We apply these models on a large dataset with ten thousands of Roman Urdu comments and a small dataset with more than two thousand comments of Urdu. Natural language constructs, English-like nature of Roman Urdu script, and Nastaleeq style of Urdu make it more challenging to process and classify the comments of both scripts using deep learning and machine learning approaches. From experiments, we find that the convolutional neural network outperforms the other models and achieves 96.2% and 91.4% accuracy on Urdu and Roman Urdu. Our results also reveal that the one-layer architectures of deep learning models give better results than two-layer architectures. Further, we compare the performance of deep learning models with five conventional machine learning models and conclude that deep learning models perform significantly better than machine learning models.

Palmprint‐based human authentication has shown great potential for civil, forensic, and corporate security applications in recent years. Palmprint recognition systems suffer because of large palmprint sizes and the presence of a large number of creases and erroneous minutiae that make the enhancement and matching phases a challenge. In this study, a novel approach is presented based on efficient enhancement and a two‐stage matching technique that demonstrates highly accurate identification results. The enhancement approach extracts minutia features from high‐quality regions based on local ridge characteristics. The selected minutiae are then matched using a two‐stage local and global minutiae neighbour‐based matching technique. To demonstrate the performance of the proposed technique, comparisons with open‐source algorithms are made based on equal error rate and detection error trade‐off graph. The results confirm the efficacy of proposed palmprint enhancement and identification technique.

Learning both hierarchical and temporal dependencies can be crucial for recurrent neural networks (RNNs) to deeply understand sequences. To this end, a unified RNN framework is required that can ease the learning of both the deep hierarchical and temporal structures by allowing gradients to propagate back from both ends without being vanished. The residual learning (RL) has appeared as an effective and less-costly method to facilitate backward propagation of gradients. The significance of the RL is exclusively shown for learning deep hierarchical representations and temporal dependencies. Nevertheless, there is lack of efforts to unify these finding into a single framework for learning deep RNNs. In this study, we aim to prove that approximating identity mapping is crucial for optimizing both hierarchical and temporal structures. We propose a framework called hierarchical and temporal residual RNNs, to learn RNNs by approximating identity mappings across hierarchical and temporal structures. To validate the proposed method, we explore the efficacy of employing shortcut connections for training deep RNNs structures for sequence learning problems. Experiments are performed on Penn Treebank, Hutter Prize and IAM-OnDB datasets and results demonstrate the utility of the framework in terms of accuracy and computational complexity. We demonstrate that even for large datasets exploiting parameters for increasing network depth can gain computational benefits with reduced size of the RNN "state".

Tuberculosis (TB) is an infectious disease that can lead towards death if left untreated. TB detection involves extraction of complex TB manifestation features such as lung cavity, air space consolidation, endobronchial spread, and pleural effusions from chest x-rays (CXRs). Deep learning based approach named convolutional neural network (CNN) has the ability to learn complex features from CXR images. The main problem is that CNN does not consider uncertainty to classify CXRs using softmax layer. It lacks in presenting the true probability of CXRs by differentiating confusing cases during TB detection. This paper presents the solution for TB identification by using Bayesian-based convolutional neural network (B-CNN). It deals with the uncertain cases that have low discernibility among the TB and non-TB manifested CXRs. The proposed TB identification methodology based on B-CNN is evaluated on two TB benchmark datasets, i.e., Montgomery and Shenzhen. For training and testing of proposed scheme we have utilized Google Colab platform which provides NVidia Tesla K80 with 12 GB of VRAM, single core of 2.3 GHz Xeon Processor, 12 GB RAM and 320 GB of disk. B-CNN achieves 96.42% and 86.46% accuracy on both dataset, respectively as compared to the state-of-the-art machine learning and CNN approaches. Moreover, B-CNN validates its results by filtering the CXRs as confusion cases where the variance of B-CNN predicted outputs is more than a certain threshold. Results prove the supremacy of B-CNN for the identification of TB and non-TB sample CXRs as compared to counterparts in terms of accuracy, variance in the predicted probabilities and model uncertainty., Bayesian-based convolutional neural network (B-CNN) for tuberculosis identification: Input chest x-ray is given to N number of CNNs. Dropout is enabled at testing stage to get posterior probability distribution (PPD) of classes.

This study presents a shallow and robust road segmentation model. The computer-aided real-time applications, like driver assistance, require real-time and accurate processing. Current studies use Deep Convolutional Neural Networks (DCNN) for road segmentation. However, DCNN requires high computational power and lots of labeled data to learn abstract features for deeper layers. The deeper the layer is, the more abstract information it tends to learn. Moreover, the prediction time of the DCNN network is an important aspect of autonomous vehicles. To overcome these issues, a Multi-feature View-based Shallow Convolutional Neural Network (MVS-CNN) is proposed that utilizes the abstract features extracted from the explicitly derived representations of the input image. Gradient information of the input image is used as additional channels to enhance the learning process of the proposed deep learning architecture. The multi-feature views are fed to a fully-connected neural network to accurately segment the road regions. The testing accuracy demonstrates that the proposed MVS-CNN achieved an improvement of 2.7% as compared to baseline CNN consisting of only RGB inputs. Furthermore, the comparison of the proposed method with the popular semantic segmentation network (SegNet) has shown that the proposed scheme performs better while being more efficient during training and evaluation. Unlike traditional segmentation techniques, which are based on the encoder-decoder architecture, the proposed MVS-CNN consists of only the encoder network. The proposed MVS-CNN has been trained and validated with two well-known datasets: the KITTI Vision Benchmark and the Cityscapes dataset. The results have been compared with the state-of-the-art deep learning architectures. The proposed MVS-CNN outperforms and shows supremacy in terms of model accuracy, processing time, and segmentation accuracy. Based on the experimental results, the proposed architecture can be considered as an efficient road segmentation architecture for autonomous vehicle systems.

Limited availability of medical imaging datasets is a vital limitation when using "data hungry" deep learning to gain performance improvements. Dealing with the issue, transfer learning has become a de facto standard, where a pre-trained convolution neural network (CNN), typically on natural images (e.g., ImageNet), is finetuned on medical images. Meanwhile, pre-trained transformers, which are self-attention-based models, have become de facto standard in natural language processing (NLP) and state of the art in image classification due to their powerful transfer learning abilities. Inspired by the success of transformers in NLP and image classification, large-scale transformers (such as vision transformer) are trained on natural images. Based on these recent developments, this research aims to explore the efficacy of pre-trained natural image transformers for medical images. Specifically, we analyze pre-trained vision transformer on CheXpert and pediatric pneumonia dataset. We use CNN standard models including VGGNet and ResNet as baseline models. By examining the acquired representations and results, we discover that transfer learning from the pre-trained vision transformer shows improved results as compared to pre-trained CNN which demonstrates a greater transfer ability of the transformers in medical imaging.

Fingerprint classification is vital for reducing the search time and computational complexity of the fingerprint identification system. The robustness of classifier relies on the strength of extracted features and the ability to deal with low-quality fingerprints. The proficiency to learn accurate features from raw fingerprint images rather than explicit feature extraction makes deep convolutional neural networks (DCNNs) attractive for fingerprint classification. The DCNNs use softmax for quantifying model confidence of a class for an input fingerprint image to make a prediction. However, the softmax probabilities are not a true representation of model confidence and often misleading in feature space that may not be represented with the available training examples. The primary goal of this study is to improve the efficacy of the fingerprint classification by dealing with false positives by employing Bayesian model uncertainty. The efficacy of the proposed method is shown through experimentations on NIST special database 4 (NIST-4) and fingerprint verification competition 2002 database 1-A (FVC DB1-A) 2002 and 2004 datasets. Results show that 0.8-1.0% of accuracy is improved with model uncertainty over the conventional DCNN.

Bandwidth constraint is one of the significant concerns of surgical telementoring, especially in rural areas. High-Efficiency Video Coding (H.265/HEVC) based video compression techniques have shown ...

A contemporary approach for acquiring the computational gains of depth in recurrent neural networks (RNNs) is to hierarchically stack multiple recurrent layers. However, such performance gains come with the cost of challenging optimization of hierarchal RNNs (HRNNs) which are deep both hierarchically and temporally. Though the researchers have exclusively highlighted the significance of using highways (or direct shortcuts) for learning deep hierarchical representations and deep temporal dependencies. However, no significant efforts are made to unify these finding into a single framework for learning deep HRNNs. We propose hierarchical recurrent highway network (HRHN) that contains highway within the hierarchical and temporal structure of the network for unimpeded information propagation across both dimensions, thus alleviating gradient vanishing problem. The proposed HRHN contain significantly reduced data-dependent parameters as compared to related methods. The experiments on language modeling (LM) tasks have demonstrated that the proposed architecture leads to design effective models. On character-level LM using Hutter prize dataset, the model achieved entropy of 2.4 bits per character. The attained perplexity on world-level LM using Penn TreeBank is 68.1. The HRHN outperformed other baseline and related models that we tested.

To find the shortest path between the nodes of a graph, different algorithms like Bellman-Ford, Dijkstra, Floyd-Warshall and Johnson exist. However, in this paper, the issue of shortest path problem with special reference to Dijkstra’s algorithm is presented. An idea of shortlisting the appropriate nodes in a graph is proposed and presented, which is then used to find the shortest path with the help of Dijkstra’s algorithm. This complete work -named Capton algorithm- provides a solution to single source shortest path problem with minimized time complexity as compared to Dijkstra’s algorithm.

A contemporary approach for acquiring the computational gains of depth in recurrent neural networks (RNNs) is to hierarchically stack multiple recurrent layers. However, such performance gains come with the cost of challenging optimization of hierarchal RNNs (HRNNs) which are deep both hierarchically and temporally. The researchers have exclusively highlighted the significance of using shortcuts for learning deep hierarchical representations and deep temporal dependencies. However, no significant efforts are made to unify these finding into a single framework for learning deep HRNNs. We propose residual recurrent highway network (R2HN) that contains highways within temporal structure of the network for unimpeded information propagation, thus alleviating gradient vanishing problem. The hierarchical structure learning is posed as residual learning framework to prevent performance degradation problem. The proposed R2HN contain significantly reduced data-dependent parameters as compared to related methods. The experiments on language modeling (LM) tasks have demonstrated that the proposed architecture leads to design effective models. On LM experiments with Penn TreeBank, the model achieved 60.3 perplexity and outperformed baseline and related models that we tested.

Automated human authentication is becoming increasingly important in today’s world due to increased need of security and surveillance applications deployed in almost all premises and installations. In this regard, palmprint biometric based identification has gained a lot of attention in recent years. However, due to large size of palmprint images and presence of principal lines, wrinkles, creases, and other noises, there are large number of inaccurate minutiae present. The computational requirement of palmprint identification is also quite large and it takes a lot of time to find identity of a palmprint in large database. In this study, a novel palmprint identification solution has been proposed that increases the accuracy of minutia detection based on improved frequency estimation and a novel region-quality based minutia extraction algorithm. Furthermore, a novel, efficient and highly accurate minutiae based encoding and matching algorithm is proposed that is designed to achieve maximum parallelism, and it is further accelerated using graphical processing unit. The results of the proposed palmprint identification demonstrate high accuracy and much faster identification speeds in comparison with current state of the art. Therefore, it can be considered as a robust, efficient and practical solution for palmprint based identification systems.

Social media has become a platform of first choice where one can express his/her feelings with freedom. The sports and matches being played are also discussed on social media such as Twitter. In this article, efforts are made to investigate the feasibility of using collective knowledge obtained from microposts posted on Twitter to predict the winner of a Cricket match. For predictions, we use three different methods that depend on the total number of tweets before the game for each team, fans sentiments toward each team and fans score predictions on Twitter. By combining these three methods, we classify winning team prediction in a Cricket game before the start of game. Our results are promising enough to be used for winning team forecast. Furthermore, the effectiveness of supervised learning algorithms is evaluated where Support Vector Machine (SVM) has shown advantage over other classifiers.

Automatic detection of heart beats constitutes the basis for electrocardiogram (ECG) analysis and mainly relies on detecting QRS complexes. Detection is typically performed by analyzing the ECG signal. However, when signal quality is low, it often leads to the triggering of false alarms. A contemporary approach to reduce false alarm rate is to use multimodal data such as arterial blood pressure (ABP) or photoplethysmogram (PPG) signals. To leverage the correlated temporal nature of these signals, a probabilistic data fusion model for heart beat detection is proposed. A hidden Markov model is used to decode waveforms into segments. A Bayesian network is employed for capturing intersegmental coupling between waveforms and detecting heart beats. The performance of the proposed system was evaluated on a dataset provided by PhysioNet Challenge 2014: Robust Detection of Heart Beats in Multimodal Data. The proposed method is comparatively analyzed with a baseline hidden Markov model method for ECG and an improvement of 9% in sensitivity and 26% in positive predictivity is observed. The efficiency of the proposed model is also compared with related data fusion methods and a comparable performance is found. The robustness of the method is analyzed by inducting Gaussian noise into the dataset. A performance gain of 31% both in sensitivity and positive predictivity is obtained in the worst case where both ECG and ABP are noisy with -6 dB signal-to-noise ratio.

Search engines and social networks are two entirely different data sources that can provide valuable information about Influenza. While search engine hosts can deliver popular queries (or terms) used for searching the Influenza related information, the social networks contain useful links of information sources that people have found valuable. The authors hypothesize that such data sources can provide vital first-hand information. In this article, they have proposed a methodology for detecting the information sources from social networks, particularly Twitter. The data filtering and source finding tasks are posed as classification tasks. Search engine queries are used for extracting related dataset. Results have shown that propose approach can be beneficial for extracting useful information regarding side effects, medications and to track geographical location of epidemics affected area.

Surgical telementoring systems have gained lots of interest, especially in remote locations. However, bandwidth constraint has been the primary bottleneck for efficient telementoring systems. This study aims to establish an efficient surgical telementoring system, where the qualified surgeon (mentor) provides real-time guidance and technical assistance for surgical procedures to the on-spot physician (surgeon). High Efficiency Video Coding (HEVC/H.265)–based video compression has shown promising results for telementoring applications. However, there is a trade-off between the bandwidth resources required for video transmission and quality of video received by the remote surgeon. In order to efficiently compress and transmit real-time surgical videos, a hybrid lossless-lossy approach is proposed where surgical incision region is coded in high quality whereas the background region is coded in low quality based on distance from the surgical incision region. For surgical incision region extraction, state-of-the-art deep learning (DL) architectures for semantic segmentation can be used. However, the computational complexity of these architectures is high resulting in large training and inference times. For telementoring systems, encoding time is crucial; therefore, very deep architectures are not suitable for surgical incision extraction. In this study, we propose a shallow convolutional neural network (S-CNN)–based segmentation approach that consists of encoder network only for surgical region extraction. The segmentation performance of S-CNN is compared with one of the state-of-the-art image segmentation networks (SegNet), and results demonstrate the effectiveness of the proposed network. The proposed telementoring system is efficient and explicitly considers the physiological nature of the human visual system to encode the video by providing good overall visual impact in the location of surgery. The results of the proposed S-CNN-based segmentation demonstrated a pixel accuracy of 97% and a mean intersection over union accuracy of 79%. Similarly, HEVC experimental results showed that the proposed surgical region–based encoding scheme achieved an average bitrate reduction of 88.8% at high-quality settings in comparison with default full-frame HEVC encoding. The average gain in encoding performance (signal-to-noise) of the proposed algorithm is 11.5 dB in the surgical region. The bitrate saving and visual quality of the proposed optimal bit allocation scheme are compared with the mean shift segmentation–based coding scheme for fair comparison. The results show that the proposed scheme maintains high visual quality in surgical incision region along with achieving good bitrate saving. Based on comparison and results, the proposed encoding algorithm can be considered as an efficient and effective solution for surgical telementoring systems for low-bandwidth networks.

The term "predatory journal" is referred for journals that exploit publishing model of open access publishers. They accept manuscripts that may either have flaws in terms of scholarly quality or they are charging fees to authors without delivering necessary editorial and pub-lishing services. To the best interest of high quality scholarly publications, the predatory jour-nals should be avoided for publications. Predatory journals are detected manually by using blacklists or using some non-verified heuristics. The former is frail since with continuously growing outbreak, it may be infeasible to constantly act as a watchdog. The latter is fragile since the heuristics are based on observations of some authors and have no documented proof of effectiveness. The paper presents a methodology and analysis to design automated predatory journal detection system. The detection task is posed as classification problem and utility of two different feature sets are tested for training classifiers. The first feature set is based on heuristics while the second set is composed of text features. Three classification algorithms are used to measure effectiveness of feature sets including kNN, support vector machine (SVM) and naive Bays. Since each classifier makes some assumptions about dataset, comparatively analyze between classifiers is an additional objective of this study. The highest achieved performance is by using SVM with heuristics-based features is 0.98 while SVM with text-based features show 0.96 which is second highest. However, one advantage that text-based features has over heuris-tics-based is that text features are comparatively easy to extract.

Efficient feature selection is an important phase of designing an effective text categorization system. Various feature selection methods have been proposed for selecting dissimilar feature sets. It is often essential to evaluate that which method is more effective for a given task and what size of feature set is an effective model selection choice. Aim of this paper is to answer these questions for designing Urdu text categorization system. Five widely used feature selection methods were examined using six well-known classification algorithms: naive Bays (NB), k-nearest neighbor (KNN), support vector machines (SVM) with linear, polynomial and radial basis kernels and decision tree (i.e. J48). The study was conducted over two test collections: EMILLE collection and a naive collection. We have observed that three feature selection methods i.e. information gain, Chi statistics, and symmetrical uncertain, have performed uniformly in most of the cases if not all. Moreover, we have found that no single feature selection method is best for all classifiers. While gain ratio out-performed others for naive Bays and J48, information gain has shown top performance for KNN and SVM with polynomial and radial basis kernels. Overall, linear SVM with any of feature selection methods including information gain, Chi statistics or symmetric uncertain methods is turned-out to be first choice across other combinations of classifiers and feature selection methods on moderate size naive collection. On the other hand, naive Bays with any of feature selection method have shown its advantage for a small sized EMILLE corpus.

this extended work, a syntactic and semantic analysis of Urdu Modal verbs is presented using lexical functional grammar encoded in an environment of the XLE parser. This grammar is developed to understand the syntactic and semantic structure of Urdu modal verbs taken from the sentences of a corpus. Based on this analysis, Urdu modal verbs are finally classified. The corpus is analyzed and sentences including modal verbs like رک اناپ (kar pAnA) 'can do', رک انکس (kar saknA) 'can do', انرک پ ڑ ا (karnA paRA) 'must/obliged to do', etc., are extracted from the corpus. Rules are studied, analyzed and encoded for these particular sentences in the form of lexical functional grammar which provides us finally the syntactic and semantic structure of different classifications.

A layered multiagent model is presented for the management of buildings by utilizing agent capabilities (i.e. reactive, proactive and adaptive behaviour) at various levels of automation within a building (i.e. at the physical, reactive, planning and adaptive level). A framework for the application of the model is also described, employing three different types of agents within a building: personal agents, building agents and a facilitator, which is mediating between the former agents to satisfy the needs of their human owners. Although the agents have different tasks and complexity, it is demonstrated that the presented model can be used for each of the various agents in a unified manner by taking advantage of the modularity of the model. An example application shows how the proposed agent typology can be applied to building automation so that both real-time tasks (e.g. negotiating the temperature of a room) and planning tasks (e.g. booking and customizing a room including a fee for service) can be realized., 8th IFAC International Conference on Fieldbuses and Networks in Industrial and Embedded Systems, May 20-22, 2009, Hanyang University, Republic of Korea

Smart HomeTechnologies are revolutionizing the ways humans interact with the world around them via digital interfaces. However these tasks are challenging at the same time. In this work, we address the issue of using complex mechanical interfaces though the medium of a user-friendly, ubiquitous, low-cost and unified digital interface, the smart phone. This is achieved through application-level programming of smart phones to recognize hand signs, which could allow mobility-restricted and elderly users to interface with their household appliances, remotely. In addition, we allow remote access for operating these appliances for flexibility and control. We prove this concept by training on visual examples of hand signs and report a correct recognition of over 88%, and test the viability of the concept under continued use by noting ordinary power consumption for uses of up to 90 minutes.

Reusability of software is found to be a key feature of quality. The most obvious outcomes of software reuse are overcoming the software crisis, advancing in software quality and improving productivity. The issue of spotting reusable software components from given existing system is very important but yet it is not much cultivated. For identification and evaluation of reusable software we use an approach that has foundation of software models and metrics. Idea of this study is to examine the competence and effectiveness of machine learning regression techniques which are experimented here to build precise and constructive evaluation model that can assess the reusability of Object Oriented based software components based on the values of five metrics of metrics suite presented by “Shyam R. Chaidmber and Chris F. Kemerer”. By setting different values of parameters of these algorithms, it is also concluded that which specific algorithm or class of algorithms is appropriate for reusability evaluation and with which parameter's values. For this comparative analysis we have used Weka and experimented different regression techniques as Multi-linear regression, Model Tree M5P, Standard instance-based learning scheme IBk and Meta-learning scheme Additive Regression. As the result of this analysis and experimentation “Standard instance-based learning IBk with no distance weighting” is found to be the best regression algorithm for reusability evaluation of Object Oriented software components using CK metrics.

In automated energy management systems, to make instantaneous decisions based on the appliance status information, continuous data access is a key requirement. With the advances in sensor and communication technologies, it is now possible to remotely monitor the power consumption data. However, before an appliance is actively monitored, it must be identified using the obtained power consumption data. Appropriate methods are required to analyse power consumption patterns for proper appliance recognition. The focus of this work is to provide the model structure for storing and distinguishing the recurring footprints of the household appliances. Hidden Markov model based method is proposed to recognize the individual appliances from combined load. It is found that the proposed method can efficiently differentiate the power consumption patterns of appliances from their combined profiles.

Microgrids are well-suited for electrification of remote off-grid areas. This paper sketches the concept of a plug-and-play microgrid with a minimum of configuration effort needed for setup. When the load of such an off-grid microgrid grows over the generation capacity and energy storage is not sufficient, demand has to be reduced to prevent a blackout. In order to decide which loads are inessential and can be shedded, automated load recognition on the basis of measured power consumption profiles is needed. Two promising approaches from the area of speech recognition, Dynamic Time Warping and Hidden Markov Models, are compared for this application. It is found that a key feature to achieve good recognition efficiency is a careful selection of the features extracted from the measured power data.

Microgrids missing a tie-line to a supporting grid may suffer from a lack of power generation. This is especially the case if they are heavily based on renewable energy resources as e.g. in remote area electrification. This paper follows a demand side management approach, where unessential loads get selectively disconnected from the grid in an under-generation scenario. In order to automatically detect unessential loads, load recognition on the basis of measured consumption data can be performed. In this article, different approaches for automatic load recognition are analyzed. It is found that especially a Hidden Markov Model based approach is suitable to fulfill the given task and provide good recognition rates. The recognition system is aimed to be part of a novel control architecture for off-grid microgrids in developing countries.

A microgrid is the power system of choice for the electrification of rural areas in developing countries. It should be able to adapt to changing load situations without the need for specialists to change the configuration of the microgrid controller. This paper proposes a self-configuring microgrid management system that is able to adjust both generation and demand of the system, so that also in case of growing electricity demand the grid can still be operable by disconnecting unessential loads. A crucial task for the microgrid controller is to automatically identify the connected loads on the basis of their consumption behaviors. For this, a template-matching algorithm is proposed that is based on Dynamic Time Warping, which is primarily used in speech recognition. It has been found that for load profile analysis, simple signal features such as the number of rising edges or the aggregated energy consumption in a given time window is sufficient to describe the signal. In contrast to speech recognition, frequency domain analysis is not necessary.

Expert systems are used extensively in many domains. This paper discusses the use of medical expert systems in Pakistan. Countries all over the world pay special attention on health facilities. A country like Pakistan faces a lot of trouble in health sector. Several attempts have been made in Pakistan to improve the health conditions of the people but the situation is still not encouraging. There is a shortage of doctors and other trained personnel in Pakistan. Expert systems can play a vital role in such cases where the medical expert is not readily available. The purpose of this paper is to analyze the role that such systems can play in improving the health conditions of the people in Pakistan., {"references":["Alison Cawsey, Introductory AI course, available at\nhttp://www.macs.hw.ac.uk/~alison/ai3notes/all.html, accessed 30/12/07.","www.pap.org.pk/Statistics.htm, accessed 06/08/26.","Assessment of Health Status and Trends in Pakistan, DRAFT June 11,\n2001, Community Medicine Residency Program, Department of\nCommunity Sciences, AKU, available at http://66.216.90.50/CHS/pdf/\nHealth Situ- ationandTrend-Pakistan,2001.pdf.","S. Siddiqi, I. U. Haq, A. Ghaffar, T. Akhtar and R. Mahaini, Pakistan-s\nMaternal and Child health Policy: Analysis, Lessons, and the Way\nForward, Health Policy 2004; 69: 117-130.","Kostas S. Metaxiotis, Emanuel Samouilidis: Expert systems in medicine:\nacademic illusion or real power? Inf. Manag. Comput. Security 8(2): 75-\n79 (2000).","http://en.wikipedia.org/wiki/Expert_system, accessed 30/12/07."]}

Fast retrieval of data has been a need of user in any database application. This paper introduces a buffer based query optimization technique in which queries are assigned weights according to their number of execution in a query bank. These queries and their optimized executed plans are loaded into the buffer at the start of the database application. For every query the system searches for a match in the buffer and executes the plan without creating new plans., {"references":["K. Shim, T. Sellis and D. Nau, Improvements on a heuristic algorithm\nfor multiple-query optimization, Data and Knowledge Engineering, 12,\n1994.","Yannis E. Ioannidis, Query Optimization. ACM Computer Surv. 28(1):\n121-123 (1996).","Georgia Kouttrika and Yannis Ionnidis, Ruled-based query\npersonalization in digital libraries, International journal digital library, 4:\n60-63 (2004).","Hristidis V, Koudas N, Papakonstantinou Y, Prefer: a system for the\nefficient execution of multiparametric ranked. Proceedings of the ACM\nSIGMOD international conference on management of data, Santa\nBarbara, CA, 21-24 May 2001. pp 259-270."]}