Feature weighting for clustering by particle swarm optimization

Clustering has been the most popular method for data exploration. Clustering is partitioning the data set into sub-partitions based on some measures say the distance measure, each partition has its own significant information. There are a number of algorithms explored for this purpose, one such algorithm is the Particle Swarm Optimization(PSO) which is a population based heuristic search technique derived from swarm intelligence. In this paper we present an improved version of the Particle Swarm Optimization where, each feature of the data set is given significance accordingly by adding some random weights, which also minimizes the distortions in the dataset if any. The performance of the above proposed algorithm is evaluated using some benchmark datasets from Machine Learning Repository. The experimental results shows that our proposed methodology performs significantly better than the previously performed experiments.

Robust formulations for clustering-based large-scale classification

Chebyshev-inequality-based convex relaxations of Chance-Constrained Programs (CCPs) are shown to be useful for learning classifiers on massive datasets. In particular, an algorithm that integrates efficient clustering procedures and CCP approaches for computing classifiers on large datasets is proposed. The key idea is to identify high density regions or clusters from individual class conditional densities and then use a CCP formulation to learn a classifier on the clusters. The CCP formulation ensures that most of the data points in a cluster are correctly classified by employing a Chebyshev-inequality-based convex relaxation. This relaxation is heavily dependent on the second-order statistics. However, this formulation and in general such relaxations that depend on the second-order moments are susceptible to moment estimation errors. One of the contributions of the paper is to propose several formulations that are robust to such errors. In particular a generic way of making such formulations robust to moment estimation errors is illustrated using two novel confidence sets. An important contribution is to show that when either of the confidence sets is employed, for the special case of a spherical normal distribution of clusters, the robust variant of the formulation can be posed as a second-order cone program. Empirical results show that the robust formulations achieve accuracies comparable to that with true moments, even when moment estimates are erroneous. Results also illustrate the benefits of employing the proposed methodology for robust classification of large-scale datasets.

Calibration of etched fiber bragg grating sensor arrays for measurement of molecular surface adsorption

Etched Fiber Bragg Grating (EFBG) sensors are attractive from the point of the inherently high multiplexing ability of fiber based sensors. However, the strong dependence of the sensitivity of EFBG sensors on the fiber diameter requires robust methods for calibration when used for distributed sensing in a large array format. Using experimental data and numerical modelling, we show that knowledge of the wavelength shift during the etch process is necessary for high-fidelity calibration of EFBG arrays. However as this approach requires the monitoring of every element of the sensor array during etching, we also proposed and demonstrated a calibration scheme using data from bulk refractometry measurements conducted post-fabrication without needing any information about the etching process. Although this approach is not as precise as the first one, it may be more practical as there is no requirement to monitor each element of the sensor array. We were able to calibrate the response of the sensors to within 3% with the approach using information acquired during etching and to within 5% using the post-fabrication bulk refractometry approach in spite of the sensitivities of the array element differing by more than a factor of 4. These two approaches present a tradeoff between accuracy and practicality.

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.

CO2 sensing at room temperature using carbon nanotubes coated core fiber Bragg grating

The sensing of carbon dioxide (CO2) at room temperature, which has potential applications in environmental monitoring, healthcare, mining, biotechnology, food industry, etc., is a challenge for the scientific community due to the relative inertness of CO2. Here, we propose a novel gas sensor based on clad-etched Fiber Bragg Grating (FBG) with polyallylamine-amino-carbon nanotube coated on the surface of the core for detecting the concentrations of CO2 gas at room temperature, in ppm levels over a wide range (1000 ppm-4000 ppm). The limit of detection observed in polyallylamine-amino-carbon nanotube coated core-FBG has been found to be about 75 ppm. In this approach, when CO2 gas molecules interact with the polyallylamine-amino-carbon nanotube coated FBG, the effective refractive index of the fiber core changes, resulting in a shift in Bragg wavelength. The experimental data show a linear response of Bragg wavelength shift for increase in concentration of CO2 gas. Besides being reproducible and repeatable, the technique is fast, compact, and highly sensitive. (C) 2013 AIP Publishing LLC.

pH Induced switching in hydrogel coated fiber bragg grating sensor

In this paper we report a novel hydrogel functionalized optical Fiber Bragg Grating (FBG) sensor based on chemo-mechanical-optical sensing, and demonstrate its specific application in pH activated process monitoring. The sensing mechanism is based on the stress due to ion diffusion and polymer phase transition which produce strain in the FBG. This results in shift in the Bragg wavelength which is detected by an interrogator system. A simple dip coating method to coat a thin layer of hydrogel on the FBG has been established. The gel consists of sodium alginate and calcium chloride. Gel formation is observed in real-time by continuously monitoring the Bragg wavelength shift. We have demonstrated pH sensing in the range of pH of 2 to 10. Another interesting phenomenon is observed by swelling and deswelling of FBG functionalized with hydrogel by a sequence of alternate dipping between acidic and base solutions. It is observed that the Bragg wavelength undergoes reversible and repeatable pH dependent switching.

Blood pressure evaluation using sphygmomanometry assisted by arterial pulse waveform detection by fiber Bragg grating pulse device

We report a blood pressure evaluation methodology by recording the radial arterial pulse waveform in real time using a fiber Bragg grating pulse device (FBGPD). Here, the pressure responses of the arterial pulse in the form of beat-to-beat pulse amplitude and arterial diametrical variations are monitored. Particularly, the unique signatures of pulse pressure variations have been recorded in the arterial pulse waveform, which indicate the systolic and diastolic blood pressure while the patient is subjected to the sphygmomanometric blood pressure examination. The proposed method of blood pressure evaluation using FBGPD has been validated with the auscultatory method of detecting the acoustic pulses (Korotkoff sounds) by an electronic stethoscope. (C) 2013 Society of Photo-Optical Instrumentation Engineers (SPIE)

Real time monitoring of petroleum leakage detection using etched fiber Bragg grating

Detection of petroleum leakages in pipelines and storage tanks is a very important as it may lead to significant pollution of the environment, accidental hazards, and also it is a very important fuel resource. Petroleum leakage detection sensor based on fiber optics was fabricated by etching the fiber Bragg grating (FBG) to a region where the total internal reflection is affected. The experiment shows that the reflected Bragg's wavelength and intensity goes to zero when etched FBG is in air and recovers Bragg's wavelength and intensity when it is comes in contact with petroleum or any external fluid. This acts as high sensitive, fast response fluid optical switch in liquid level sensing, petroleum leakage detection etc. In this paper we present our results on using this technique in petroleum leakage detection.

Carbon nanotube coated fiber Bragg grating for photomechanical optic modulator

We have demonstrated novel concept of utilizing the photomechanical actuation in carbon nanotubes (CNTs) to tune and reversibly switch the Bragg wavelength. When fiber Bragg grating coated with CNTs (CNT-FBG) is exposed externally to a wide range of optical wavelengths, e. g., ultraviolet to infrared (0.2-200 mu m), a strain is induced in the CNTs which alters the grating pitch and refractive index in the CNT-FBG system resulting in a shift in the Bragg wavelength. This novel approach will find applications in telecommunication, sensors and actuators, and also for real time monitoring of the photomechanical actuation in nanoscale materials. (C) 2013 AIP Publishing LLC.

Learning from positive and unlabelled examples using maximum margin clustering

Learning from Positive and Unlabelled examples (LPU) has emerged as an important problem in data mining and information retrieval applications. Existing techniques are not ideally suited for real world scenarios where the datasets are linearly inseparable, as they either build linear classifiers or the non-linear classifiers fail to achieve the desired performance. In this work, we propose to extend maximum margin clustering ideas and present an iterative procedure to design a non-linear classifier for LPU. In particular, we build a least squares support vector classifier, suitable for handling this problem due to symmetry of its loss function. Further, we present techniques for appropriately initializing the labels of unlabelled examples and for enforcing the ratio of positive to negative examples while obtaining these labels. Experiments on real-world datasets demonstrate that the non-linear classifier designed using the proposed approach gives significantly better generalization performance than the existing relevant approaches for LPU.

Modified particle swarm optimization for pattern clustering

Data clustering is a common technique for statistical data analysis, which is used in many fields, including machine learning and data mining. Clustering is grouping of a data set or more precisely, the partitioning of a data set into subsets (clusters), so that the data in each subset (ideally) share some common trait according to some defined distance measure. In this paper we present the genetically improved version of particle swarm optimization algorithm which is a population based heuristic search technique derived from the analysis of the particle swarm intelligence and the concepts of genetic algorithms (GA). The algorithm combines the concepts of PSO such as velocity and position update rules together with the concepts of GA such as selection, crossover and mutation. The performance of the above proposed algorithm is evaluated using some benchmark datasets from Machine Learning Repository. The performance of our method is better than k-means and PSO algorithm.

Simultaneous feature selection and clustering using particle swarm optimization

Data clustering groups data so that data which are similar to each other are in the same group and data which are dissimilar to each other are in different groups. Since generally clustering is a subjective activity, it is possible to get different clusterings of the same data depending on the need. This paper attempts to find the best clustering of the data by first carrying out feature selection and using only the selected features, for clustering. A PSO (Particle Swarm Optimization)has been used for clustering but feature selection has also been carried out simultaneously. The performance of the above proposed algorithm is evaluated on some benchmark data sets. The experimental results shows the proposed methodology outperforms the previous approaches such as basic PSO and Kmeans for the clustering problem.

Support vector clustering-based direct coherency identification of generators in a multi-machine power system

This study investigates the application of support vector clustering (SVC) for the direct identification of coherent synchronous generators in large interconnected multi-machine power systems. The clustering is based on coherency measure, which indicates the degree of coherency between any pair of generators. The proposed SVC algorithm processes the coherency measure matrix that is formulated using the generator rotor measurements to cluster the coherent generators. The proposed approach is demonstrated on IEEE 10 generator 39-bus system and an equivalent 35 generators, 246-bus system of practical Indian southern grid. The effect of number of data samples and fault locations are also examined for determining the accuracy of the proposed approach. An extended comparison with other clustering techniques is also included, to show the effectiveness of the proposed approach in grouping the data into coherent groups of generators. This effectiveness of the coherent clusters obtained with the proposed approach is compared in terms of a set of clustering validity indicators and in terms of statistical assessment that is based on the coherency degree of a generator pair.

Fiber Bragg Grating Sensor Instrumentation for Parking Space Occupancy Management

This work aims at providing an effective parking management system by reducing the drivers' searching time for vacant car-parking space, in turn improving the traffic flow in the car park areas. This is achieved by the use of Fiber Bragg Grating Sensor (FBG) sensor instrumentation in vehicle parking management system. Present work involves embedding an array of FBG sensors underground in the parking space, then determining the strain changes on the FBG sensor due to load applied by the vehicle parked in the parking space, occupancy of the parking space is determined. To validate the FBG sensor parking management system, three most common cases have been considered. This closed loop FBG parking management system can give real-time feed-back to space-guidance display board helping the driver in maneuvering the vehicle to the appropriate parking space. The proposed technique offers optimized usage of parking space for the various segments of cars and also facilitates in a conjoined automated billing system, as compared to conventional method of parking systems.

Design and development of Fiber Bragg Grating sensing plate for plantar strain measurement and postural stability analysis

This paper describes an ab initio design and development of a novel Fiber Bragg Grating (FBG) sensor based strain sensing plate for the measurement of plantar strain distribution in human foot. The primary aim of this work is to study the feasibility of usage of FBG sensors in the measurement of plantar strain in the foot; in particular, to spatially resolve the strain distribution in the foot at different regions such as fore-foot, mid-foot and hind-foot. This study also provides a method to quantify and compare relative postural stability of different subjects under test; in addition, traditional accelerometers have been used to record the movements of center of gravity (second lumbar vertebra) of the subject and the results obtained have been compared against the outcome of the postural stability studies undertaken using the developed FBG plantar strain sensing plate. (C) 2013 Elsevier Ltd. All rights reserved.