This research paper aims at exploiting efficient ways of implementing the N-Body problem. The N-Body problem, in the field of physics, predicts the movements and planets and their gravitational interactions. In this paper, the efficient execution of heavy computational work through usage of different cores in CPU and GPU is looked into; achieved by integrating the OpenMP parallelization API and the Nvidia CUDA into the code. The paper also aims at performance analysis of various algorithms used to solve the same problem. This research not only aids as an alternative to complex simulations but also for bigger data that requires work distribution and computationally expensive procedures.
In most corruption scandals, the use of front companies for money laundering is almost ubiquitous. This work proposes to apply image classification to detect such organizations, through the use of Convolutional Neural Networks (CNN), namely the AlexNet architecture. The images are obtained by address search in Google Street View API, and the resulting classification will be further used along with other features to detect front com- panies in order to help the auditors from the Ministry of Transparency and Office of the Comptroller General (CGU, in Portuguese). To this moment, we applied classification to almost 15 thousand suppliers scenes with active contracts with the Brazilian Government until September 2016, obtained through data matching between the Government Purchases database and the Brazilian Federal Revenue Office database (more recent scenes should be added as this work progresses). Preliminary results with a pre-trained AlexNet CNN show the need for developing new scene classes more suited to the Brazilian context. In order to do this, we propose to apply clustering algorithms in features extracted from the last fully-connected layer of this net. The classes obtained will be used to fine-tune the AlexNet CNN for future classification, through the use of training from scratch or fine tuning techniques.
The decreasing of groundwater quality has been the major issue in Tangerang area. One of the key process is the interaction between groundwater and Cisadane river water, which flows over volcanic deposits of Bojongmanik Fm, Genteng Fm, Tuf Banten, and Alluvial Fan. The objective of this study is to unravel such interactions based on the potentiometric mapping in the riverbank. We had 60 stop sites along the riverbank for groundwater and river water level observations, and chemical measurements (TDS, EC, temp, and pH). Three river water gauge were also analyzed to see the fluctuations.
We identified three types of hydrodynamic relationships with fairly low flow gradients: effluent flow at Segmen I (Kranggan - Batuceper) with 0.2-0.25 gradient, perched flow at Segmen II (Batuceper-Kalibaru) with gradient 0.2-0.25, and influent flow at Segmen III (Kalibaru-Tanjungburung) with gradient 0.15-0.20. Such low flow gradient is controlled by the moderate to low morphological slope in the area. The gaining and losing stream model were also supported by the river water fluctuation data. TDS and EC readings increased more than 40% from upstream to downstream. At some points the both measurements were two times higher than the permissible limits, along with the drops of pH values at those areas.
This study shows the very close interaction between Cisadane river water and groundwater in the riverbank. Therefore the authorities need to be managed the areas with a very strict regulations related to the small and large scale industries located near by the river.
Dasapta Erwin Irawan, Deny Juanda Puradimaja, Defitri Yeni, Arno Adi Kuntoro, Miga Magenika Julian