We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to .
To save content items to your Kindle, first ensure coreplatform@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.
Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
Find out more about the Kindle Personal Document Service.
Spectral clustering is a technique for finding group structure in data. It makes use of the spectrum of the data similarity matrix to perform dimensionality reduction for clustering in fewer dimensions. Spectral clustering algorithms have been shown to be more effective in finding clusters than traditional algorithms such as k-means. However, spectral clustering suffers from a scalability problem in both memory use and computation time when the size of a dataset is large. To perform clustering on large datasets, in this work, we parallelize both memory use and computation using MapReduce and MPI. Through an empirical study on a document set of 534,135 instances and a photo set of 2,121,863 images, we show that our parallel algorithm can effectively handle large problems.
Clustering is one of the most important subfields of machine learning and data mining tasks. In the last decade, spectral clustering (e.g., Shi and Malik, 2000; Meila and Shi, 2000; Fowlkes et al., 2004), motivated by normalized graph cut, has attracted much attention. Unlike traditional partition-based clustering, spectral clustering exploits a pairwise data similarity matrix. It has been shown to be more effective than traditional methods such as k-means, which considers only the similarity between instances and k centroids (Ng, Jordan, and Weiss, 2001).Because of its effectiveness, spectral clustering has been widely used in several areas such as information retrieval and computer vision (e.g., Dhillon, 2001; Xu, Liu, and Gong, 2003; Shi and Malik, 2000; Yu and Shi, 2003).
Non-alloyed ohmic contacts using Ti/Pt/Au and Ni/Ge/Au on InGaAs/GaAs layers grown by Molecular Beam Epitaxy (MBE) have been investigated. The n-type InGaAs film has a doping concentration higher than 1X1019 cm-3. Specific contact resistance below 2X10-7 Ωcm2 could be easily achieved with Ti/Pt/Au. Due to the layer intermixing and outdiffusion of In and Ga, the specific contact resistance and sheet resistance increase after thermal treatment. When Ni/Ge/Au is used as the contact metal, the outdiffusion of In and Ga atoms is more severe than that of Ti/Pt/Au. After annealing at 450°C for two minutes, the Au4In formed and the characteristics of the contact became worse. All the phenomena illustrated above have been observed and investigated by Transmission Line Model, X-ray diffraction, Auger Electron Spectroscopy and Secondary Ion Mass Spectrum. As far as the thermal stability is concerned, it is convinced that Ti/Pt/Au is the best one of these two non-alloyed ohmic contact studied.
The first study of the TiW nitrides (TiWNx) as the Schottky contact metals to the n type Ga0.51In0 49P has been made. The Ga0.51 In0.49P epitaxial layer was successfully grown on the GaAs substrate by LP-MOCVD to form a lattice-matched heterostructure. The RF-magnetron sputtering system was utilized for the nitride deposition. The thermal stability of the nitride films were studied using rapid thermal annealing (RTA) method. Both the electrical characteristics and the materials characteristics were investigated. The materials properties of the nitride films were characterized by X-ray diffraction (XRD), Transmission electron microscopy (TEM), and Auger electron spectroscopy (AES). The TiWNx Schottky contacts demonstrate excellent electrical and physical characteristics, even after high temperature annealing. The barrier heights range from 0.81 to 1.05 eV depending on the content of the nitrogen and the annealing conditions. The XRD and AES results show no indication of interaction at the TiWNX/GaInP interface of both as-deposited and annealed samples. The outstanding characteristics of the contact were attributed to the high bandgap nature of the Ga0.51In0.49P and the incorporation of nitrogen into the TiW films.
Thermal stability of the Schottky contacts on Ga0.51In0.49P has been made. The Ga0.51In0.49P epitaxial layer was successfully grown on the GaAs substrate by LP-MOCVD to form a lattice-matched heterostructure. In this paper, materials aspects of the Ga0.51In0.49P layers were characterized and thermal stability of three different types of films, including single-layer metal (Pt, Ni, Pd, Au, Co, Mo, W, Cr, Ti, Al, Ta, and In), metal silicides (WSi2, W5Si3, PtSi, and Pt2i), and TiW nitrides (TiWNx ) as the Schottky contacts materials on Ga0.51In0.49P were studied. Due to the high bandgap nature of Ga0.51In0.49P, the Schottky contacts on Ga0.51In0.49P demonstrate good characteristics. The barrier heights range from 0.79 to 1.19 eV depending on the selection of the materials and the annealing conditions. For single-metal contacts, Pt film shows the best thermal stability, the barrier height of 1.09 eV and the ideality factor of 1.06 were obtained for the Pt Schottky diode with furnace annealing at 500 °C for 30 min. For refractory compound films, the TiWNx film shows the best thermal stability. The TiWNx Schottky contacts demonstrate excellent electrical as well as physical characteristics, even after high temperature annealing at 850°C.
The passivation of GaAs MESFET's with PECVD silicon nitride films of both compressive and tensile stresses is reported. The shift of MESFET parameters due to passivation was found to be dependent upon gate orientation. For example, it is found that the shifts of threshold voltage are of opposite sign for MESFET's oriented along [011] and [011] directions. Our experiments show that nitride of tensile stress is preferable for MESFET's with [011] oriented gates. The shifts in VTH IDSS and GM of the devices before and after nitride passivation are lesd thaw s% if the nitride of appropriate stress states are used for passivation.
Breakdown voltage of the MESFET's after nitride deposition was also studied. It is found that the process with higher hydrogen incorporation tends to reduce the surface oxide and increase the breakdown voltage after nitride deposition.
Email your librarian or administrator to recommend adding this to your organisation's collection.