Online suspicious event detection in a constrained environment with RGB+D camera using multi-stream CNNs and SVM

Multimedia Tools and Applications - Automated detection of human actions is very important for surveillance of government offices, Bank-ATMs, etc. to prevent crime and loss of property and lives of...     

Analysing stable feature selection through an augmented marine predator algorithm based on opposition-based learning

Retrieving the relevant information from the high-dimensional dataset enhances the classification accuracy of a predictive model. This research critique has devised an improved marine predator algorithm based on opposition learning for stable feature selection to overcome the problem of high-dimensionality. Marine predator algorithm is a population-based meta-heuristics optimization algorithm that works on the ‘survival-of-the-fittest’ theory. Classical marine predator algorithm explores the search space merely in one direction, affecting its converging capacity while being responsible for stagnation at local minima. The proposed opposition-based learning nuances enhance the exploration capacity of marine predator algorithm and productively converges the model to global optima. The proposed OBL-based marine predator algorithm selects stable, substantial elements from six different high-dimensional microarray datasets. The performance of the proposed method is investigated using five predominantly used classifiers. From the result, it is understood that the proposed approach outperforms other conventional feature selection techniques in terms of converging capability, classification accuracy, and stable feature selection.     

Improved salp swarm algorithm based on the levy flight for feature selection

The Journal of Supercomputing - The fields of data science and data mining are enduring high-dimensionality issues because of a high volume of data. Conventional machine learning techniques give...     

Affordable housing finance companies in India: how do they ‘differently’ serve the underserved?

Abstract

Over the years, the importance of access to affordable finance has inevitably been recognized as a critical component of developing the third world. Against the backdrop of huge under-penetration of the housing finance sector in India, this study reflects on the functioning of Affordable Housing Finance Companies (AHFCs), which are among the important stakeholders providing affordable housing loans to the low-income segment households. However, the process through which these new entrants have been able to down-market housing finance remains a black box till date. In light of this research gap, this study proposes a conceptual model that succinctly captures the business process of AHFCs across three main dimensions: Outreach Approach, Lending and Underwriting Practices and Risk Management Interventions. This model reflects on the kaleidoscope of process innovations that the AHFCs have embraced to cater to the housing finance needs of the low-income customers, while achieving profitability and social impact.     

Kinetics of hydrogenation of 4‐chloro‐2‐nitrophenol catalyzed by Pt/carbon catalyst

Hydrogenation of 4‐chloro‐2‐nitrophenol (CNP) was carried out at moderate hydrogen pressures, 7–28 atm, and temperatures in the range 298–313 K using Pt/carbon and Pd/γ‐Al₂O₃ as catalysts in a stirred pressure reactor. Hydrogenation of CNP under the above conditions gave 4‐chloro‐2‐aminophenol (CAP). Dechlorination to form 2‐aminophenol and 2‐nitrophenol is observed when hydrogenation of CNP is carried out above 338 K, particularly with Pd/γ‐Al₂O₃ catalyst. Among the catalysts tested, 1%Pt/C was found to be an effective catalyst for the hydrogenation of CNP to form CAP, exclusively. To confirm the absence of gas–liquid mass transfer effects on the reaction, the effect of stirring speed (200–1000 rpm) and catalyst loading (0.02–0.16 g) on the initial reaction rate at maximum temperature 310 K and substrate concentration (0.25 mole) were thoroughly studied. The kinetics of hydrogenation of CNP carried out using 1%Pt/C indicated that the initial rates of hydrogenation had first order dependence with respect to substrate, catalyst and hydrogen pressure in the range of concentrations varied. From the Arrhenius plot of ln rate vs 1000/T, an apparent activation energy of 22 kJ mol^?1 was estimated. © 2001 Society of Chemical Industry     

Role of doping-induced photochemical and microstructural properties in the photocatalytic activity of InVO4 for splitting of water

We report in this paper on microstructural, optical and photocatalytic properties of single-phase indium orthovanadates, as a function of doping at lattice sites. The UV–visible spectra of these samples exhibited intense UV-region bands at 250 and 350 nm, besides broad absorption band in visible region (350–700 nm). The wavelength at absorption edge and the intensity of visible absorption showed considerable increase on doping of an impurity, particularly at V or O lattice sites. Also, the samples gave rise to blue-green photoluminescence emission, with overriding bands at ca. 420, 450, 460 and 485 nm, on excitation at 240–420 nm wavelengths. The intensity of these fluorescence bands varied with excitation wavelength and impurity content of a sample. In deviation with several earlier studies, only oxygen and no hydrogen were produced during photocatalytic splitting of water, in the experiments conducted under visible light (>395 nm) and at a pH of ~6.5. The O₂ yield depended on the dispersed metal co-catalyst, impurity content and the addition of methanol as sacrificial reagent. On the other hand, small quantities of hydrogen and no oxygen were evolved on UV-irradiation of pure water using metal/InVO₄. These results are ascribed to flat band potentials and the doping-induced inter-band donor and acceptor charge trapping states of InVO₄, the presence of which is revealed by XRD, luminescence and XPS studies. Our study also confirms that the onset of absorption edge may not necessarily correspond to band-to-band energy gap of a semiconducting material. This accounts for some anomalous band gap energies reported earlier for InVO₄.