Unsupervised feature selection for outlier detection in categorical data using mutual information

Outlier detection in high dimensional categorical data has been a problem of much interest due to the extensive use of qualitative features for describing the data across various application areas. Though there exist various established methods for dealing with the dimensionality aspect through feature selection on numerical data, the categorical domain is actively being explored. As outlier detection is generally considered as an unsupervised learning problem due to lack of knowledge about the nature of various types of outliers, the related feature selection task also needs to be handled in a similar manner. This motivates the need to develop an unsupervised feature selection algorithm for efficient detection of outliers in categorical data. Addressing this aspect, we propose a novel feature selection algorithm based on the mutual information measure and the entropy computation. The redundancy among the features is characterized using the mutual information measure for identifying a suitable feature subset with less redundancy. The performance of the proposed algorithm in comparison with the information gain based feature selection shows its effectiveness for outlier detection. The efficacy of the proposed algorithm is demonstrated on various high-dimensional benchmark data sets employing two existing outlier detection methods.

Employing denaturation for rapid electrochemical detection of myoglobin using TiO2 nanotubes

An alternative antibody-free strategy for the rapid electrochemical detection of cardiac myoglobin has been demonstrated here using hydrothermally synthesized TiO2 nanotubes (Ti-NT). The denaturant induced unfolding of myoglobin led to easy access of the deeply buried electroactive heme center and thus the efficient reversible electron transfer from protein to electrode surface. The sensing performance of the Ti-NT modified electrodes were compared vis a vis commercially available titania and GCEs. The tubular morphology of the Ti-NT led to facile transfer of electrons to the electrode surface, which eventually provided a linear current response (obtained from cyclic voltammetry) over a wide range of Mb concentration. The sensitivity of the Ti-NT based sensor was remarkable and was equal to 18 mu A mg(-1) ml (detection limit = 50 nM). This coupled with the rapid analysis time of a few tens of minutes (compared to a few days for ELISA) demonstrates its potential usefulness for the early detection of acute myocardial infarction (AMI).

Detection limit of etched fiber bragg grating sensors

While Fiber Bragg Grating (FBG) sensors have been extensively used for temperature and strain sensing, clad etched FBGs (EFBGs) have only recently been explored for refractive index sensing. Prior literature in EFBG based refractive index sensing predominantly deals with bulk refractometry only, where the Bragg wavelength shift of the sensor as a function of the bulk refractive index of the sample can be analytically modeled, unlike the situation for adsorption of molecular thin films on the sensor surface. We used a finite element model to calculate the Bragg wavelength change as a function of thickness and refractive index of the adsorbing molecular layer and compared the model with the real-time, in-situ measurement of electrostatic layer-by-layer (LbL) assembly of weak polyelectrolytes on the silica surface of EFBGs. We then used this model to calculate the layer thickness of LbL films and found them to be in agreement with literature. Further, we used this model to arrive at a realistic estimate of the limit of detection of EFBG sensors based on nominal measurement noise levels in current FBG interrogation systems and found that sufficiently thinned EFBGs can provide a competitive platform for real-time measurement of molecular interactions while simultaneously leveraging the high multiplexing capabilities of fiber optics.

Resonance raman detection and estimation in the aqueous phase using water dispersible cyclodextrin: reduced-graphene oxide sheets

Resonance Raman spectroscopy is a powerful analytical tool for detecting and identifying analytes, but the associated strong fluorescence background severely limits the use of the technique. Here, we show that by attaching beta-cyclodextrin (beta-CD) cavities to reduced graphene-oxide (rGO) sheets we obtain a water dispersible material (beta-CD: rGO) that combines the hydrophobicity associated with rGO with that of the cyclodextrin cavities and provides a versatile platform for resonance Raman detection. Planar aromatic and dye molecules that adsorb on the rGO domains and nonplanar molecules included within the tethered beta-CD cavities have their fluorescence effectively quenched. We show that it is possible using the water dispersible beta-CD: rGO sheets to record the resonance Raman spectra of adsorbed and included organic chromophores directly in aqueous media without having to extract or deposit on a substrate. This is significant, as it allows us to identify and estimate organic analytes present in water by resonance Raman spectroscopy.

Nonparametric distributed sequential detection via universal source coding

We consider nonparametric or universal sequential hypothesis testing when the distribution under the null hypothesis is fully known but the alternate hypothesis corresponds to some other unknown distribution. These algorithms are primarily motivated from spectrum sensing in Cognitive Radios and intruder detection in wireless sensor networks. We use easily implementable universal lossless source codes to propose simple algorithms for such a setup. The algorithms are first proposed for discrete alphabet. Their performance and asymptotic properties are studied theoretically. Later these are extended to continuous alphabets. Their performance with two well known universal source codes, Lempel-Ziv code and KT-estimator with Arithmetic Encoder are compared. These algorithms are also compared with the tests using various other nonparametric estimators. Finally a decentralized version utilizing spatial diversity is also proposed and analysed.

Efficient method of moving shadow detection and vehicle classification

Moving shadow detection and removal from the extracted foreground regions of video frames, aim to limit the risk of misconsideration of moving shadows as a part of moving objects. This operation thus enhances the rate of accuracy in detection and classification of moving objects. With a similar reasoning, the present paper proposes an efficient method for the discrimination of moving object and moving shadow regions in a video sequence, with no human intervention. Also, it requires less computational burden and works effectively under dynamic traffic road conditions on highways (with and without marking lines), street ways (with and without marking lines). Further, we have used scale-invariant feature transform-based features for the classification of moving vehicles (with and without shadow regions), which enhances the effectiveness of the proposed method. The potentiality of the method is tested with various data sets collected from different road traffic scenarios, and its superiority is compared with the existing methods. (C) 2013 Elsevier GmbH. All rights reserved.