Description:
Most existing model-based approaches to anomaly detection construct a profile of normal instances, then identify instances that do not conform to the normal profile as anomalies. This paper proposes a fundamentally different model-based method that explicitly isolates anomalies instead of profiles normal points. To our best knowledge, the concept of isolation has not been explored in current literature. The use of isolation enables the proposed method, iForest, to exploit sub-sampling to an extent that is not feasible in existing methods, creating an algorithm which has a linear time complexity with a low constant and a low memory requirement. Our empirical evaluation shows that iForest performs favourably to ORCA, a near-linear time complexity distance-based method, LOF and random forests in terms of AUC and processing time, and especially in large data sets. iForest also works well in high dimensional problems which have a large number of irrelevant attributes, and in situations where training set does not contain any anomalies.
Description:
In the last few years, there have been many works in the area of hybrid neural learning algorithms combining a global and local based method for training artificial neural networks. In this paper, we discuss various connection strategies that can be applied to a special kind of a hybrid neural learning algorithm group, one that combines a genetic algorithm-based method with various least square-based methods like QR factorization. The relative advantages and disadvantages of the different connection types are studied to find a suitable connection topology for combining the two different learning methods. The methodology also finds the optimum number of hidden neurons using a hierarchical combination methodology structure for weights and architecture. We have tested our proposed approach on XOR, 10 bit odd parity, and some other real-world benchmark data sets, such as the hand-writing character dataset from CEDAR, Breast cancer, and Heart Disease from the UCI machine learning repository.
Description:
A gene regulatory network (GRN) represents a set of genes and its regulatory interactions. The inference of the regulatory interactions between genes is usually carried out using an appropriate mathematical model and the available gene expression profile. Among the various models proposed for GRN inference, our recently proposed Michaelis–Menten based ODE model provides a good trade-off between the computational complexity and biological relevance. This model, like other known GRN models, also uses an evolutionary algorithm for parameter estimation. Considering various issues associated with such population based stochastic optimization approaches (e.g. diversity, premature convergence due to local optima, accuracy, etc.), it becomes important to seed the initial population with good individuals which are closer to the optimal solution. In this paper, we exploit the inherent strength of principal component analysis (PCA) in a novel manner to initialize the population for GRN optimization. The benefit of the proposed method is validated by reconstructing in silico and in vivo networks of various sizes. For the same level of accuracy, the approach with PCA based initialization shows improved convergence speed.
Description:
A gene regulatory network (GRN) represents a set of genes and its regulatory interactions. The inference of the regulatory interactions between genes is usually carried out using an appropriate mathematical model and the available gene expression profile. Among the various models proposed for GRN inference, our recently proposed Michaelis–Menten based ODE model provides a good trade-off between the computational complexity and biological relevance. This model, like other known GRN models, also uses an evolutionary algorithm for parameter estimation. Considering various issues associated with such population based stochastic optimization approaches (e.g. diversity, premature convergence due to local optima, accuracy, etc.), it becomes important to seed the initial population with good individuals which are closer to the optimal solution. In this paper, we exploit the inherent strength of principal component analysis (PCA) in a novel manner to initialize the population for GRN optimization. The benefit of the proposed method is validated by reconstructing in silico and in vivo networks of various sizes. For the same level of accuracy, the approach with PCA based initialization shows improved convergence speed.
Description:
Pattern-based video coding (PVC) has already established its superiority over recent video coding standard H.264, at low bit rate because of an extra pattern-mode to segment out the arbitrary shape of the moving region within the macroblock (MB). To determine the pattern-mode, the PVC however uses three thresholds to reduce the number of MBs coded using the pattern- mode. By setting these content-sensitive thresholds to any predefined values, the technique risks ignoring some MBs that would otherwise be selected by the rate-distortion optimization function for this mode. Consequently, the ultimate achievable performance is sacrificed to save motion estimation times. In this paper, a novel PVC scheme is proposed by removing all thresholds to determine this mode and hence more efficient performance is achieved without knowing the content of the video sequences. To keep computational complexity in check, pattern motion is approximated from the motion vector of the MB. In addition, efficient pattern similarity metric and new Lagrangian multipliers are also developed. The experimental results confirm that this new scheme improves the image quality by at least 0.5 dB and 1.0 dB compared to the existing PVC and the H.264 respectively
Description:
Computational complexity in data mining is attributed to algorithms but lies hugely with the data. Different algorithms may exist to solve the same problem, but the simplest is not always the best. At the same time, data of astronomical proportions is rather common, boosted by automation, and the fuller the data, the better resolution of the concept it projects. Paradoxically, it is the computing power that is lacking. Perhaps a fast algorithm can be run on the data, but not the optimal. Even then any modeling is much constrained, involving serial application of many algorithms. The only other way to relieve the computational load is via making the data lighter. Any representative subset has to preserve the data essence suiting, ideally, any algorithm. The reduction should minimize the error of approximation, while trading precision for performance. Data mining is a wide field. We concentrate on classification. In the literature review we present a variety of methods, emphasizing the effort of past decade. Two major objects of reduction are instances and attributes. The data can be also recast into a more economical format. We address sampling, noise reduction, class domain binarization, feature ranking, feature subset selection, feature extraction, and also discretization of continuous features. Achievements are tremendous, but so are possibilities. We improve an existing technique of data cleansing and suggest a way of data condensing as the extension. We also touch on noise reduction. Instance similarity, excepting the class mix, prompts a technique of feature selection. Additionally, we consider multivariate discretization, enabling a compact data representation without the size change. We compare proposed methods with alternative techniques which we introduce new, implement or use available.
Description:
Computational complexity in data mining is attributed to algorithms but lies hugely with the data. Different algorithms may exist to solve the same problem, but the simplest is not always the best. At the same time, data of astronomical proportions is rather common, boosted by automation, and the fuller the data, the better resolution of the concept it projects. Paradoxically, it is the computing power that is lacking. Perhaps a fast algorithm can be run on the data, but not the optimal. Even then any modeling is much constrained, involving serial application of many algorithms. The only other way to relieve the computational load is via making the data lighter. Any representative subset has to preserve the data essence suiting, ideally, any algorithm. The reduction should minimize the error of approximation, while trading precision for performance. Data mining is a wide field. We concentrate on classification. In the literature review we present a variety of methods, emphasizing the effort of past decade. Two major objects of reduction are instances and attributes. The data can be also recast into a more economical format. We address sampling, noise reduction, class domain binarization, feature ranking, feature subset selection, feature extraction, and also discretization of continuous features. Achievements are tremendous, but so are possibilities. We improve an existing technique of data cleansing and suggest a way of data condensing as the extension. We also touch on noise reduction. Instance similarity, excepting the class mix, prompts a technique of feature selection. Additionally, we consider multivariate discretization, enabling a compact data representation without the size change. We compare proposed methods with alternative techniques which we introduce new, implement or use available.
Description:
An academic team is a highly cohesive collaboration group of scholars, which has been recognized as an effective way to improve scientific output in terms of both quality and quantity. However, the high staff turnover brings about a series of problems that may have negative influences on team performance. To address this challenge, we first detect the tendency of the member who may potentially leave. Here, the outlierness is defined with respect to familiarity, which is quantified by using collaboration intensity. It is assumed that if a team member has a higher familiarity with scholars outside the team, then this member might probably leave the team. To minimize the influence caused by the leaving of such an outlier member, we propose an optimization solution to find a proper candidate who can replace the outlier member. Based on random walk with graph kernel, our solution involves familiarity matching, skill matching, as well as structure matching. The proposed approach proves to be effective and outperforms existing methods when applied to computer science academic teams.
Description:
An academic team is a highly cohesive collaboration group of scholars, which has been recognized as an effective way to improve scientific output in terms of both quality and quantity. However, the high staff turnover brings about a series of problems that may have negative influences on team performance. To address this challenge, we first detect the tendency of the member who may potentially leave. Here, the outlierness is defined with respect to familiarity, which is quantified by using collaboration intensity. It is assumed that if a team member has a higher familiarity with scholars outside the team, then this member might probably leave the team. To minimize the influence caused by the leaving of such an outlier member, we propose an optimization solution to find a proper candidate who can replace the outlier member. Based on random walk with graph kernel, our solution involves familiarity matching, skill matching, as well as structure matching. The proposed approach proves to be effective and outperforms existing methods when applied to computer science academic teams.
Description:
High Efficiency Video Coding (HEVC) standard adopts several new approaches to achieve higher coding efficiency (approximately 50% bit-rate reduction) compared to its predecessor H.264/AVC with same perceptual image quality. Huge computational time has also increased due to the algorithmic complexity of HEVC compared to H.264/AVC. However, it is really a demanding task to reduce the encoding time while preserving the similar quality of the video sequences. In this paper, we propose a novel efficient intermode selection technique and incorporate into HEVC framework to predict motion estimation and motion compensation modes between current and reference blocks and perform faster inter mode selection based on three dissimilar motion types in divergent video sequences. Instead of exploring and traversing all the modes exhaustively, we merely select a subset of candidate modes and the final mode from the selected subset is determined based on their lowest Lagrangian cost function. The experimental results reveal that average encoding time can be downscaled by 40% with similar rate-distortion performance compared to the exhaustive mode selection strategy in HEVC.
Description:
Blind Source Separation techniques are widely used in the field of wireless communication for a very long time to extract signals of interest from a set of multiple signals without training data. In this paper, we investigate the problem of separation of the human voice from a mixture of human voice and sounds from different musical instruments. The human voice may be a singing voice in a song or may be a part of some news, broadcast by a channel with background music. This paper proposes a generalized Short Time Fourier Transform (STFT)-based technique, combined with filter bank to extract vocals from background music. The main purpose is to design a filter bank and to eliminate background aliasing errors with best reconstruction conditions, having approximated scaling factors. Stereo signals in time-frequency domain are used in experiments. The input stereo signals are processed in the form of frames and passed through the proposed STFT-based technique. The output of the STFT-based technique is passed through the filter bank to minimize the background aliasing errors. For reconstruction, first an inverse STFT is applied and then the signals are reconstructed by the OverLap-Add method to get the final output, containing vocals only. The experiments show that the proposed approach performs better than the other state-of-the-art approaches, in terms of Signal-to-Interference Ratio (SIR) and Signal-to-Distortion Ratio (SDR), respectively.
Description:
Vendor selection is a strategic issue in supply chain management for any organization to identify the right supplier. Such selection in most cases is based on the analysis of some specific criteria. Most of the researches so far concentrate on multi-criteria decision-making analysis. Though many approaches have been proposed, analytic hierarchy process (AHP) is the most well known as it can deal with a very complex criteria structure. In AHP, the selected criteria are ranked and organized in a hierarchical order from generic to specific to formulate the problem. Though this order of ranking is acceptably logical, it incurs a huge computational complexity when a large number of alternatives are considered as the selection criteria. Moreover, the AHP may generate wrong selection due to computational error. To address these limitations, a novel model namely vendor selection using fuzzy c-means algorithm and analytic hierarchy process (VFA) is presented in this paper by integrating the fuzzy c-means clustering (FCM) algorithm with analytic hierarchy process (AHP). The outcome of the proposed VFA algorithm is compared with the basic AHP algorithm and VFA outperforms the basic AHP and reduces the computational complexity of AHP by a factor of 7.