Gait biometrics which concern with recognizing individuals by the way they walk are of a paramount importance these days. Human gait is a candidate pathway for such identification tasks since other mechanisms can be concealed. Most common methodologies rely on analyzing 2D/3D images captured by surveillance cameras. Thus, the performance of such methods depends heavily on the quality of the images and the appearance variations of individuals. In this study, we describe how gait biometrics could be used in individuals' identification using a deep feature learning and inertial measurement unit (IMU) technology. We propose a model that recognizes the biological and physical characteristics of individuals, such as gender, age, height, and weight, by examining high-level representations constructed during its learning process. The effectiveness of the proposed model has been demonstrated by a set of experiments with a new gait dataset generated using a shoe-type based on a gait analysis sensor system. The experimental results show that the proposed model can achieve better identification accuracy than existing models, while also demonstrating more stable predictive performance across different classes. This makes the proposed model a promising alternative to current image-based modeling.

The emergence of new technologies for the fifth/sixth generation (5G/6G) wireless networks has led to the development of new services, resulting in an increase in malicious activities and cyber-attacks targeting various networklayers. Edge computing, a crucial technology enabler for 6G, is expected to facilitate traffic optimisation and support new ultra-low latency services. By integrating computing power from supercomputing servers into devices at the network edge in a distributed manner, edge computing can provide consistent quality-of-service, even in remote areas, which will drive the growth of associated applications. However, the complex environment created by edge computing also poses challenges for detecting malware. Therefore, this paper proposes a novel approach to malware detection using explain ability via visualization and a multi-labelling technique. An object detection algorithm is used to identify malware families within the dataset which is created by emphasizing key regions. Using features from different malware categories in an image, this model displays a thorough malware recipe. Our experiments using real malware data demonstrate that identifying malware by its visible characteristics can significantly improve the interpretability of the detection process, enhancing transparency and trustworthiness.

This study makes time-series-based predictions on future returns of the STOXX Europe 600 and the German DAX by adopting (in addition to a lagged and transformed version of the target series) a diversified set of predictors. Feature engineering expands further — from the initial raw group of variables, to extract knowledge of market conditions and demand for hedging. A penalisation factor is introduced with loss functions to learn a model from neural networks, in order to adapt a traditional machine learning regression framework to solve the equity forecasting problems in question. Architectures based on convolutional neural network are proposed, treating the obtained feature map similarly to an image. Experiments over different time periods demonstrate that trading strategies derived from the forecasts are more profitable than models based on efficient market assumptions. The temporal, non-stationary structure of financial data has a significant impact on the out of sample success of any model. It thus can be seen that different architectures exhibit different resilience to changing market conditions.

In an increasingly interconnected world, the sophistication of cyber-attacks is on the rise. Cybersecurity research stands as a pivotal factor in shaping the prosperity of nations. To counter threats to network infrastructure and sensitive data, a multitude of security solutions with varying degrees of efficacy have been proposed. However, these solutions have thus far insufficiently accounted for a critical dimension: sustainability. In this context, sustainability entails the continuous support of processes over time by enhancing the computational requisites, scalability, energy efficiency, and resource utilization of defence and security systems. This special issue endeavors to explore recent strides in model development, innovative methodologies, and insightful observations aimed at enhancing cybersecurity, with a particular emphasis on the sustainability of defence and security systems.

Over the past decade, keystroke-based pattern recognition techniques, as a forensic tool for behavioral biometrics, have gained increasing attention. Although a number of machine learning-based approaches have been proposed, they are limited in terms of their capability to recognize and profile a set of an individual's characteristics. In addition, up to today, their focus was primarily gender and age, which seem to be more appropriate for commercial applications (such as developing commercial software), leaving out from research other characteristics, such as the educational level. Educational level is an acquired user characteristic, which can improve targeted advertising, as well as provide valuable information in a digital forensic investigation, when it is known. In this context, this paper proposes a novel machine learning model, the randomized radial basis function network, which recognizes and profiles the educational level of an individual who stands behind the keyboard. The performance of the proposed model is evaluated by using the empirical data obtained by recording volunteers' keystrokes during their daily usage of a computer. Its performance is also compared with other well-referenced machine learning models using our keystroke dynamic datasets. Although the proposed model achieves high accuracy in educational level prediction of an unknown user, it suffers from high computational cost. For this reason, we examine ways to reduce the time that is needed to build our model, including the use of a novel data condensation method, and discuss the tradeoff between an accurate and a fast prediction. To the best of our knowledge, this is the first model in the literature that predicts the educational level of an individual based on the keystroke dynamics information only.

The present contribution proposes a spectrally efficient censor-based cooperative spectrum sensing (C-CSS) approach in a sustainable cognitive radio network that consists of multiple antenna nodes and experiences imperfect sensing and reporting channels. In this context, exact analytic expressions are first derived for the corresponding probability of detection, probability of false alarm, and secondary throughput, assuming that each secondary user (SU) sends its detection outcome to a fusion center only when it has detected a primary signal. Capitalizing on the findings of the analysis, the effects of critical measures, such as the detection threshold, the number of SUs, and the number of employed antennas, on the overall system performance are also quantified. In addition, the optimal detection threshold for each antenna based on the Neyman-Pearson criterion is derived and useful insights are developed on how to maximize the system throughput with a reduced number of SUs. It is shown that the C-CSS approach provides two distinct benefits compared with the conventional sensing approach, i.e., without censoring: i) the sensing tail problem, which exists in imperfect sensing environments, can be mitigated; and ii) less SUs are ultimately required to obtain higher secondary throughput, rendering the system more sustainable.

Recent studies have proposed that traditional security technology -- involving pattern-matching algorithms that check predefined pattern sets of intrusion signatures -- should be replaced with sophisticated adaptive approaches that combine machine learning and behavioural analytics. However, machine learning is performance driven, and the high computational cost is incompatible with the limited computing power, memory capacity and energy resources of portable IoT-enabled devices. The convoluted nature of deep-structured machine learning means that such models also lack transparency and interpretability. The knowledge obtained by interpretable learners is critical in security software design. We therefore propose two novel models featuring a common Deep Extraction and Mutual Information Selection (DEMISe) element which extracts features using a deep-structured stacked autoencoder, prior to feature selection based on the amount of mutual information (MI) shared between each feature and the class label. An entropy-based tree wrapper is used to optimise the feature subsets identified by the DEMISe element, yielding the DEMISe with Tree Evaluation and Regression Detection (DETEReD) model. This affords 'white box' insight, and achieves a time to build of 603 seconds, a 99.07% detection rate, and 98.04% model accuracy. When tested against AWID, the best-referenced intrusion detection dataset, the new models achieved a test error comparable to or better than state-of-the-art machine-learning models, with a lower computational cost and higher levels of transparency and interpretability.

We propose a spectrally efficient censor-based cooperative spectrum sensing (C-CSS) approach for a sustainable cognitive radio network that consists of multiple antenna nodes and experiences imperfect sensing and reporting channels. First, analytic expressions are derived for the corresponding probabilities of detection and false alarm, assuming that each secondary user sends its detection outcome to a fusion center only when it believes to have detected a primary user's signal. Second, we derive lower bounds for the probability of false alarm, where we show that a sensing tail problem, which exist in the conventional (non-censor-based) scheme, can be effectively mitigated with the aid of the proposed C-CSS scheme. Simulation results are presented to corroborate the derived analytic results, and to provide theoretical and technical insights that are useful for the design of cognitive radio networks.

Low-level criminals, who do the legwork in a criminal organization, are the most likely to be arrested, whereas the high-level ones tend to avoid attention. But crippling the work of criminal organizations is not possible unless investigators can identify the most influential, high-level members and monitor their communication channels. Investigators often approach this task by requesting the mobile phone service records of the arrested low-level criminals to identify contacts, and then they build a network model of the organization, where each node denotes a criminal and the edges represent communications. Network analysis can be used to infer the most influential criminals and most important communication channels within the network, but screening all the nodes and links in a network is laborious and time consuming. Here, we propose a new forensic analysis system called identifying influential criminals and their communication channels (IICCC) that can effectively and efficiently infer the high-level criminals and short-list the important communication channels in a criminal organization, based on the mobile phone communications of its members. IICCC can also be used to build a network from crime incident reports. We evaluated IICCC experimentally and compared it with five other systems, confirming its superior prediction performance.

The recent advances in mobile technologies have resulted in Internet of Things (IoT)-enabled devices becoming more pervasive and integrated into our daily lives. The security challenges that need to be overcome mainly stem from the open nature of a wireless medium, such as a Wi-Fi network. An impersonation attack is an attack in which an adversary is disguised as a legitimate party in a system or communications protocol. The connected devices are pervasive, generating high-dimensional data on a large scale, which complicates simultaneous detections. Feature learning, however, can circumvent the potential problems that could be caused by the large-volume nature of network data. This paper thus proposes a novel deep-feature extraction and selection (D-FES), which combines stacked feature extraction and weighted feature selection. The stacked autoencoding is capable of providing representations that are more meaningful by reconstructing the relevant information from its raw inputs. We then combine this with modified weighted feature selection inspired by an existing shallow-structured machine learner. We finally demonstrate the ability of the condensed set of features to reduce the bias of a machine learner model as well as the computational complexity. Our experimental results on a well-referenced Wi-Fi network benchmark data set, namely, the Aegean Wi-Fi Intrusion data set, prove the usefulness and the utility of the proposed D-FES by achieving a detection accuracy of 99.918% and a false alarm rate of 0.012%, which is the most accurate detection of impersonation attacks reported in the literature.

Smartphones have become a vital part of our business and everyday life, as they constitute the primary communication vector. Android dominates the smartphone market (86.2%) and has become pervasive, running in `smart' devices such as tablets, TV, watches, etc. Nowadays, instant messaging applications have become popular amongst smartphone users and since 2016 are the main way of messaging communication. Consequently, their inclusion in any forensics analysis is necessary as they constitute a source of valuable data, which might be used as (admissible) evidence. Often, their examination involves the extraction and analysis of the applications' databases that reside in the device's internal or external memory. The downfall of this method is the fact that databases can be tampered or erased, therefore the evidence might be accidentally or maliciously modified. In this paper, a methodology for retrieving instant messaging data from the volatile memory of Android smartphones is proposed, instead of the traditional database retrieval. The methodology is demonstrated with the use of a case study of four experiments, which provide insights regarding the behavior of such data in memory. Our experimental results show that a large amount of data can be retrieved from the memory, even if the device's battery is removed for a short time. In addition, the retrieved data are not only recent messages, but also messages sent a few months before data acquisition.

We introduce a forensic analysis system called ECLfinder that identifies the influential members of a criminal organization as well as the immediate leaders of a given list of lower-level criminals. Criminal investigators usually seek to identify the influential members of criminal organizations, because eliminating them is most likely to hinder and disrupt the operations of these organizations and put them out of business. First, ECLfinder constructs a network representing a criminal organization from either mobile communication data associated with the organization or crime incident reports that include information about the organization. It then constructs a minimum spanning tree (MST) of the network. It identifies the influential members of a criminal organization by determining the important vertices in the network representing the organization, using the concept of existence dependence. Each vertex v is assigned a score, which is the number of other vertices, whose existence in MST is dependent on v. Vertices are ranked based on their scores. Criminals represented by the top ranked vertices are considered the influential members of the criminal organization represented by the network. We evaluated the quality of ECLfinder by comparing it experimentally with three other systems. Results showed marked improvement.

BACKGROUND: All proteins associate with other molecules. These associated molecules are highly predictive of the potential functions of proteins. The association of a protein and a molecule can be determined from their co-occurrences in biomedical abstracts. Extensive semantically related co-occurrences of a protein's name and a molecule's name in the sentences of biomedical abstracts can be considered as indicative of the association between the protein and the molecule. Dependency parsers extract textual relations from a text by determining the grammatical relations between words in a sentence. They can be used for determining the textual relations between proteins and molecules. Despite their success, they may extract textual relations with low precision. This is because they do not consider the semantic relationships between terms in a sentence (i.e., they consider only the structural relationships between the terms). Moreover, they may not be well suited for complex sentences and for long-distance textual relations. RESULTS: We introduce an information extraction system called PPFBM that predicts the functions of unannotated proteins from the molecules that associate with these proteins. PPFBM represents each protein by the other molecules that associate with it in the abstracts referenced in the protein's entries in reliable biological databases. It automatically extracts each co-occurrence of a protein-molecule pair that represents semantic relationship between the pair. Towards this, we present novel semantic rules that identify the semantic relationship between each co-occurrence of a protein-molecule pair using the syntactic structures of sentences and linguistics theories. PPFBM determines the functions of an un-annotated protein p as follows. First, it determines the set S r of annotated proteins that is semantically similar to p by matching the molecules representing p and the annotated proteins. Then, it assigns p the functional category FC if the significance of the frequency of occurrences of S r in abstracts associated with proteins annotated with FC is statistically significantly different than the significance of the frequency of occurrences of S r in abstracts associated with proteins annotated with all other functional categories. We evaluated the quality of PPFBM by comparing it experimentally with two other systems. Results showed marked improvement. CONCLUSIONS: The experimental results demonstrated that PPFBM outperforms other systems that predict protein function from the textual information found within biomedical abstracts. This is because these system do not consider the semantic relationships between terms in a sentence (i.e., they consider only the structural relationships between the terms). PPFBM's performance over these system increases steadily as the number of training protein increases. That is, PPFBM's prediction performance becomes more accurate constantly, as the size of training proteins gets larger. This is because every time a new set of test proteins is added to the current set of training proteins. A demo of PPFBM that annotates each input Yeast protein (SGD (Saccharomyces Genome Database). Available at: http://www.yeastgenome.org/download-data/curation) with the functions of Gene Ontology terms is available at: (see Appendix for more details about the demo) http://ecesrvr.kustar.ac.ae:8080/PPFBM/.

Upon publication of the original article [1] it was found that author corrections for Table 1 were not correctly implemented. You can now find the correct Table 1 below and also updated in the original article. Table 1 The distribution of semantically related and semantically unrelated co-occurrences of molecule m i and Protein p Pair in an Abstract A j

Botnets, which consist of remotely controlled compromised machines called bots, provide a distributed platform for several threats against cyber world entities and enterprises. Intrusion detection system (IDS) provides an efficient countermeasure against botnets. It continually monitors and analyzes network traffic for potential vulnerabilities and possible existence of active attacks. A payload-inspection-based IDS (PI-IDS) identifies active intrusion attempts by inspecting transmission control protocol and user datagram protocol packet's payload and comparing it with previously seen attacks signatures. However, the PI-IDS abilities to detect intrusions might be incapacitated by packet encryption. Traffic-based IDS (T-IDS) alleviates the shortcomings of PI-IDS, as it does not inspect packet payload; however, it analyzes packet header to identify intrusions. As the network's traffic grows rapidly, not only the detection-rate is critical, but also the efficiency and the scalability of IDS become more significant. In this paper, we propose a state-of-the-art T-IDS built on a novel randomized data partitioned learning model (RDPLM), relying on a compact network feature set and feature selection techniques, simplified subspacing and a multiple randomized meta-learning technique. The proposed model has achieved 99.984% accuracy and 21.38 s training time on a well-known benchmark botnet dataset. Experiment results demonstrate that the proposed methodology outperforms other well-known machine-learning models used in the same detection task, namely, sequential minimal optimization, deep neural network, C4.5, reduced error pruning tree, and randomTree.

Members of a criminal organization, who hold central positions in the organization, are usually targeted by criminal investigators for removal or surveillance. This is because they play key and influential roles by acting as commanders, who issue instructions or serve as gatekeepers. Removing these central members (i.e., influential members) is most likely to disrupt the organization and put it out of business. Most often, criminal investigators are even more interested in knowing the portion of these influential members, who are the immediate leaders of lower level criminals. These lower level criminals are the ones who usually carry out the criminal works; therefore, they are easier to identify. The ultimate goal of investigators is to identify the immediate leaders of these lower level criminals in order to disrupt future crimes. We propose, in this paper, a forensic analysis system called SIIMCO that can identify the influential members of a criminal organization. Given a list of lower level criminals in a criminal organization, SIIMCO can also identify the immediate leaders of these criminals. SIIMCO first constructs a network representing a criminal organization from either mobile communication data that belongs to the organization or crime incident reports. It adopts the concept space approach to automatically construct a network from crime incident reports. In such a network, a vertex represents an individual criminal, and a link represents the relationship between two criminals. SIIMCO employs formulas that quantify the degree of influence/importance of each vertex in the network relative to all other vertices. We present these formulas through a series of refinements. All the formulas incorporate novel-weighting schemes for the edges of networks. We evaluated the quality of SIIMCO by comparing it experimentally with two other systems. Results showed marked improvement.

The influential members of a criminal organization are usually targeted by criminal investigators for removal or surveillance. Identifying and capturing these influential members will most likely to disrupt the organization. We propose in this paper a forensic analysis system called CLDRI that can identify the most influential members of a criminal organization. First, a network representing a criminal organization is built from Mobile Communication Data that belongs to the organization. In such a network, a vertex represents an individual criminal and an edge represents the communication attempts between two criminals. CLDRI employs formulas that quantify the degree of importance of each vertex in the network relative to all other vertices. We present these formulas through series of improvement refinements. All the formulas incorporate novel-weighting schemes for the edges of networks. We evaluated the quality of CLDRI by comparing it experimentally with two systems. Results showed improvement.

We present a Natural Language Processing extraction system called IESforPFP, which can retrieve useful information from biomedical abstracts. IESforPFP aims at enhancing the state of the art of biological text mining by applying novel linguistic computational technique. By retrieving significant patterns of associations between proteins and molecules from biomedical abstracts, IESforPFP can determine the functions of un-annotated proteins. The system determines the semantic relationship between each protein-molecule pair in sentences using novel semantic rules. It applies a semantic relationship extraction model that retrieves information from different structural forms of constituents in sentences. In the framework of IESforPFP, each protein p is represented by a vector of weights. Each weight reflects the significance of a molecule m in the biomedical abstracts associated with p. That is, each weight quantifies the likelihood of the association between m and p. IESforPFP determines the set of annotated proteins that is semantically similar to p by comparing their vectors. It then annotates p with the functions of these annotated proteins. We evaluated the quality of IESforPFP by comparing it experimentally with two other systems. Results showed marked improvement.

Proline residues are common source of kinetic complications during folding. The X-Pro peptide bond is the only peptide bond for which the stability of the cis and trans conformations is comparable. The cis-trans isomerization (CTI) of X-Pro peptide bonds is a widely recognized rate-limiting factor, which can not only induces additional slow phases in protein folding but also modifies the millisecond and sub-millisecond dynamics of the protein. An accurate computational prediction of proline CTI is of great importance for the understanding of protein folding, splicing, cell signaling, and transmembrane active transport in both the human body and animals. In our earlier work, we successfully developed a biophysically motivated proline CTI predictor utilizing a novel tree-based consensus model with a powerful metalearning technique and achieved 86.58 percent Q2 accuracy and 0.74 Mcc, which is a better result than the results (70-73 percent Q2 accuracies) reported in the literature on the well-referenced benchmark dataset. In this paper, we describe experiments with novel randomized subspace learning and bootstrap seeding techniques as an extension to our earlier work, the consensus models as well as entropy-based learning methods, to obtain better accuracy through a precise and robust learning scheme for proline CTI prediction.

We propose a classifier system called iPFPi that predicts the functions of un-annotated proteins. iPFPi assigns an un-annotated protein P the functions of GO annotation terms that are semantically similar to P. An un-annotated protein P and a GO annotation term T are represented by their characteristics. The characteristics of P are GO terms found within the abstracts of biomedical literature associated with P. The characteristics of Tare GO terms found within the abstracts of biomedical literature associated with the proteins annotated with the function of T. Let F and F/ be the important (dominant) sets of characteristic terms representing T and P, respectively. iPFPi would annotate P with the function of T, if F and F/ are semantically similar. We constructed a novel semantic similarity measure that takes into consideration several factors, such as the dominance degree of each characteristic term t in set F based on its score, which is a value that reflects the dominance status of t relative to other characteristic terms, using pairwise beats and looses procedure. Every time a protein P is annotated with the function of T, iPFPi updates and optimizes the current scores of the characteristic terms for T based on the weights of the characteristic terms for P. Set F will be updated accordingly. Thus, the accuracy of predicting the function of T as the function of subsequent proteins improves. This prediction accuracy keeps improving over time iteratively through the cumulative weights of the characteristic terms representing proteins that are successively annotated with the function of T. We evaluated the quality of iPFPi by comparing it experimentally with two recent protein function prediction systems. Results showed marked improvement.

Wafer defects, which are primarily defective chips on a wafer, are of the key challenges facing the semiconductor manufacturing companies, as they could increase the yield losses to hundreds of millions of dollars. Fortunately, these wafer defects leave unique patterns due to their spatial dependence across wafer maps. It is thus possible to identify and predict them in order to find the point of failure in the manufacturing process accurately. This paper introduces a novel simplified subspaced regression framework for the accurate and efficient identification of defect patterns in semiconductor wafer maps. It can achieve a test error comparable to or better than the state-of-the-art machine-learning (ML)-based methods, while maintaining a low computational cost when dealing with large-scale wafer data. The effectiveness and utility of the proposed approach has been demonstrated by our experiments on real wafer defect datasets, achieving detection accuracy of 99.884% and R2of 99.905%, which are far better than those of any existing methods reported in the literature.

Biologists may need to know the set of genes that are semantically related to a given set of genes. For instance, a biologist may need to know the set of genes related to another set of genes known to be involved in a specific disease. Some works use the concept of gene clustering in order to identify semantically related genes. Others propose tools that return the set of genes that are semantically related to a given set of genes. Most of these gene similarity measures determine the semantic similarities among the genes based solely on the proximity to each other of the GO terms annotating the genes, while overlook the structural dependencies among these GO terms, which may lead to low recall and precision of results. We propose in this paper a search engine called IRG, which overcomes the problems of current gene similarity measures outlined above. The search engine constructs a minimum spanning tree of GO graph based on their weights. Let S' be the set of genes that are semantically related to set S. In the framework of IRG, the set S' is annotated to the GO term located at the convergence of the subtree of the minimum spanning tree that passes through the GO terms annotating the set S. We evaluated IRG experimentally and compared it with a gene prediction tool called DynGO and with two other systems we proposed previously. Results showed marked improvement.

BACKGROUND: Protein chains are generally long and consist of multiple domains. Domains are distinct structural units of a protein that can evolve and function independently. The accurate prediction of protein domain linkers and boundaries is often regarded as the initial step of protein tertiary structure and function predictions. Such information not only enhances protein-targeted drug development but also reduces the experimental cost of protein analysis by allowing researchers to work on a set of smaller and independent units. In this study, we propose a novel and accurate domain-linker prediction approach based on protein primary structure information only. We utilize a nature-inspired machine-learning model called Random Forest along with a novel domain-linker profile that contains physiochemical and domain-linker information of amino acid sequences. RESULTS: The proposed approach was tested on two well-known benchmark protein datasets and achieved 68% sensitivity and 99% precision, which is better than any existing protein domain-linker predictor. Without applying any data balancing technique such as class weighting and data re-sampling, the proposed approach is able to accurately classify inter-domain linkers from highly imbalanced datasets. CONCLUSION: Our experimental results prove that the proposed approach is useful for domain-linker identification in highly imbalanced single- and multi-domain proteins.

Proline cis-trans isomerization (CTI) plays a key role in the rate-determining steps of protein folding. Accurate prediction of proline CTI is of great importance for the understanding of protein folding, splicing, cell signaling, and transmembrane active transport in both the human body and animals. Our goal is to develop a state-of-the-art proline CTI predictor based on a biophysically motivated intelligent consensus modeling through the use of sequence information only (i.e., position specific scores generated by PSI-BLAST). The current computational proline CTI predictors reach about 70-73 percent Q2 accuracies and about 0.40 Matthew correlation coefficient (Mcc) through the use of sequence-based evolutionary information as well as predicted protein secondary structure information. However, our approach that utilizes a novel decision tree-based consensus model with a powerful randomized-metal earning technique has achieved 86.58 percent Q2 accuracy and 0.74 Mcc, on the same proline CTI data set, which is a better result than those of any existing computational proline CTI predictors reported in the literature.

Data sampling is a widely used technique in a broad range of machine learning problems. Traditional sampling approaches generally rely on random resampling from a given dataset. However, these approaches do not take into consideration additional information, such as sample quality and usefulness. We recently proposed a data sampling technique, called sample subset optimization (SSO). The SSO technique relies on a cross-validation procedure for identifying and selecting the most useful samples as subsets. In this paper, we describe the application of SSO techniques to imbalanced and ensemble learning problems, respectively. For imbalanced learning, the SSO technique is employed as an under-sampling technique for identifying a subset of highly discriminative samples in the majority class. In ensemble learning, the SSO technique is utilized as a generic ensemble technique where multiple optimized subsets of samples from each class are selected for building an ensemble classifier. We demonstrate the utilities and advantages of the proposed techniques on a variety of bioinformatics applications where class imbalance, small sample size, and noisy data are prevalent.

A new paradigm of thinking pertaining to a novel holistic intelligent campus (iCampus) environment is proposed, in this paper, in order to enrich and enhance, as well as to transform, the end-to-end learning lifecycle of a knowledge ecosystem. Analogous to the different functions of a biological brain, the central digital nervous system of the campus is comprised of various different interconnected functional intelligences. Each of these intelligent areas is set to perform its specified functional role in a dynamic and coherent inter- and intra-integrative manner within the environment itself. A generalized roadmap has also been devised to encapsulate the concept from within an existing or a new campus setting. Note that the nature of the iCampus proposition is inherently multi-disciplinary and has multi-applicability to other forms of intelligent environment. To capture part of the essence of the concept, some of the key challenges pertinent to the iCampus ecosystem have also been highlighted within the campus value proposition framework.

In this study, we propose a new machine learning model namely, Adaptive Locality-Effective Kernel Machine (Adaptive-LEKM) for protein phosphorylation site prediction. Adaptive-LEKM proves to be more accurate and exhibits a much stable predictive performance over the existing machine learning models. Adaptive-LEKM is trained using Position Specific Scoring Matrix (PSSM) to detect possible protein phosphorylation sites for a target sequence. The performance of the proposed model was compared to seven existing different machine learning models on newly proposed PS-Benchmark_1 dataset in terms of accuracy, sensitivity, specificity and correlation coefficient. Adaptive-LEKM showed better predictive performance with 82.3% accuracy, 80.1% sensitivity, 84.5% specificity and 0.65 correlation-coefficient than contemporary machine learning models.

BACKGROUND: Post-translational modifications have a substantial influence on the structure and functions of protein. Post-translational phosphorylation is one of the most common modification that occur in intracellular proteins. Accurate prediction of protein phosphorylation sites is of great importance for the understanding of diverse cellular signalling processes in both the human body and in animals. In this study, we propose a new machine learning based protein phosphorylation site predictor, SiteSeek. SiteSeek is trained using a novel compact evolutionary and hydrophobicity profile to detect possible protein phosphorylation sites for a target sequence. The newly proposed method proves to be more accurate and exhibits a much stable predictive performance than currently existing phosphorylation site predictors. RESULTS: The performance of the proposed model was compared to nine existing different machine learning models and four widely known phosphorylation site predictors with the newly proposed PS-Benchmark_1 dataset to contrast their accuracy, sensitivity, specificity and correlation coefficient. SiteSeek showed better predictive performance with 86.6% accuracy, 83.8% sensitivity, 92.5% specificity and 0.77 correlation-coefficient on the four main kinase families (CDK, CK2, PKA, and PKC). CONCLUSION: Our newly proposed methods used in SiteSeek were shown to be useful for the identification of protein phosphorylation sites as it performed much better than widely known predictors on the newly built PS-Benchmark_1 dataset.

BACKGROUND: Protein domains present some of the most useful information that can be used to understand protein structure and functions. Recent research on protein domain boundary prediction has been mainly based on widely known machine learning techniques, such as Artificial Neural Networks and Support Vector Machines. In this study, we propose a new machine learning model (IGRN) that can achieve accurate and reliable classification, with significantly reduced computations. The IGRN was trained using a PSSM (Position Specific Scoring Matrix), secondary structure, solvent accessibility information and inter-domain linker index to detect possible domain boundaries for a target sequence. RESULTS: The proposed model achieved average prediction accuracy of 67% on the Benchmark_2 dataset for domain boundary identification in multi-domains proteins and showed superior predictive performance and generalisation ability among the most widely used neural network models. With the CASP7 benchmark dataset, it also demonstrated comparable performance to existing domain boundary predictors such as DOMpro, DomPred, DomSSEA, DomCut and DomainDiscovery with 70.10% prediction accuracy. CONCLUSION: The performance of proposed model has been compared favourably to the performance of other existing machine learning based methods as well as widely known domain boundary predictors on two benchmark datasets and excels in the identification of domain boundaries in terms of model bias, generalisation and computational requirements.