564 research outputs found
Feature Selection with Non Linear PCA: A Neural Network Approach
No full textMachine learning consists in the creation and development of algorithms that allow a machine to learn itself, gradually improving its behavior over time. This learning is more effective, the more representative is the features of the dataset used to describe the problem. An important objective is therefore the correct selection (and, possibly, reduction of the number) of the most rele- vant features, which is typically carried out through dimensional reduction tools such as Principal Component Analysis (PCA), which is not linear in the more general case. In this work, an approach to the calculation of the reduced space of the PCA is proposed through the definition and implementation of appropriate models of artificial neural network, which allows to obtain an accurate and at the same time flexible reduction of the dimensionality of the problem
Luminescence Investigation of Direct and Indirect Excitons Bound to Deep-Neutral Acceptors in e-GaSe
No full textWe report on our investigations of excitonic luminescence spectra of Bridgmann grown GaSe crystals from the liquid N2 temperature up to 300 K and at weak laser excitation intensity. We measured the spontaneous luminescence due to direct and indirect excitonic recombinations. Moreover, a detailed analysis of the intensity of the lines composing the emission spectrum versus the temperature and the excitation intensity permits to detect and assign two lines of the excitonic
luminescence as due to the recombination of direct as well as indirect excitons bound to localized centres in the forbidden energy gap. On the basis of recent theories on bound exciton-impurity complexes we found that both direct and indirect excitons are bound to neutral and deep acceptor centres. From the measured dissociation energy of these bound excitons we could estimate the ionization energy of the acceptor levels, which were already identified independently in GaSe by electrical transport measurements
Kinetics of Radiative Recombinations in GaSe and Influence of Cu doping on the Luminescence Spectra
No full textSpontaneous photoluminescence (PL) spectra of Cu-doped and undoped -GaSe have been investigated in the temperature range from 80 to 300 K and at low laser-excitation intensity (P) from 10-3 to 10 W cm-2. The main modification of the spectra in doped crystals with respect to those of undoped samples is the appearance of two bands in the extrinsic part of the PL spectrum and centered at 655 and 678 nm, respectively. The luminescence at energies below the excitonic recombinations (extrinsic bands) is enhanced by doping. Also indirect free- and bound-excitonic lines are also strongly influenced by the impurity concentration; in fact, their emission intensity, which depends linearly on P in undoped crystals, shows a quadratic dependence in doped samples. The temperature dependence of the PL spectra gives the thermal activation energy of some extrinsic bands, which results equal the ionization energy of the acceptor levels involved in the extrinsic transitions. A simple kinetic model of the radiative recombination is proposed; it accounts for the experimental data of the excitation intensity dependence of the free- and bound-excitonic lines. This model can also explain the different temperature dependence of the PL intensity of these lines: linear for the free-excitonic emissions, exponential for the bound-excitonic recombinations. Some radiative transitions from donor levels located in the energy gap of GaSe are analyzed and a scheme of donor and acceptor states involved in the PL spectra is proposed
Direct and Indirect Excitonic Emission in GaSe
No full textPhotoluminescence spectra of undoped crystals of the layer semiconductor GaSe have been measured from 80 K up to room temperature. The direct and indirect excitonic emissions are clearly observed in the intrinsic part of the spectrum. Moreover, a detailed analysis of the luminescence intensity has been made as a function of exciting power and temperature, allowing one to ascribe some lines of the intrinsic part of emission spectrum to direct and indirect excitons bound to localized impurity levels in the energy gap of GaSe
Thermalization of Photoexcited Localized Excitons in GaSe Samples with Stacking Disorder
No full textWe have studied the photoluminescence of GaSe at 80 K under energy-selective excitation conditions. For excitation energies on the low-energy side of the n=1 direct free-exciton absorption line we find that the emission line due to the n=1 direct free excitons follows rigidly the excitation energy, whereas for excitation energies on the high-energy side the respective emission spectrum shifts to lower energies and the linewidth increases. Similar behavior is found on exciting into the excited n=2 direct exciton states. At room temperature and for excitation energies larger than the direct gap the emission spectra become independent of the excitation energy. We show that the experimental findings can be understood in terms of an extended version of the multiple-trapping model, which accounts for the localization of excitons in the direction perpendicular to the layers of GaSe. This localization is a consequence of the stacking disorder present in our samples. The resulting physical picture is that at low temperatures and for low excitation energies thermalization effects can be neglected within the recombination lifetime, whereas at high temperatures or high excitation energies the thermalization takes place within the recombination lifetime
Dynamics of Excitons in GaSe as a Function of Temperature
No full textThe dynamics of direct and indirect excitons, either free or localized by the layer stacking disorder is studied in GaSe at temperatures from 2 to 150 K and as a fünction of the excitation energy
A Simulated Annealing Algorithm for Scheduling Problems
No full textAn algorithm using the heuristic technique of Simulated Annealing to solve a scheduling problem is presented, focusing on the scheduling issues. The ap- proximated method is examined together with its key parameters (freezing, tempering, cooling, number of contours to be explored), and the choices made in identifying these parameters are illustrated to generate a good algorithm that efficiently solves the scheduling problem
A Wafer Bin Map "Relaxed" Clustering Algorithm for Improving Semiconductor Production Yield
No full textThe semiconductor manufacturing process involves long and complex activities, with intensive use of resources. Producers compete through the introduction of new technologies for increasing yield and reducing costs. So, yield improvement is becoming increasingly important since advanced production technologies are complex and interrelated. In particular, Wafer Bin Maps (WBMs) presenting specific fault models provide crucial information to keep track of process problems in semiconductor manufacturing. Production control is often based on the "judgement" of expert engineers who, however, carry out the analysis of map templates through simple visual exploration. In this way, existing studies are subjective, time consuming, and are also limited by the capacity of human recognition. This study proposes a network-based data mining approach, which integrates correlation graphs with clustering analysis to quickly extract patterns from WBMs and then bind them to manufacturing defects.
An empirical study has been conducted on real production data for validating the proposed clustering algorithm, which showed a perfect correspondence between the malfunction patterns found by the algorithm and those discovered by human experts, so confirming the validity of our approach in its ability of correctly identifying actual defective patterns to help improving production yield
- …
