Application of K-tree to document clustering

Digital collections are growing exponentially in size as the information age takes a firm grip on all aspects of society. As a result Information Retrieval (IR) has become an increasingly important area of research. It promises to provide new and more effective ways for users to find information relevant to their search intentions. Document clustering is one of the many tools in the IR toolbox and is far from being perfected. It groups documents that share common features. This grouping allows a user to quickly identify relevant information. If these groups are misleading then valuable information can accidentally be ignored. There- fore, the study and analysis of the quality of document clustering is important. With more and more digital information available, the performance of these algorithms is also of interest. An algorithm with a time complexity of O(n2) can quickly become impractical when clustering a corpus containing millions of documents. Therefore, the investigation of algorithms and data structures to perform clustering in an efficient manner is vital to its success as an IR tool. Document classification is another tool frequently used in the IR field. It predicts categories of new documents based on an existing database of (doc- ument, category) pairs. Support Vector Machines (SVM) have been found to be effective when classifying text documents. As the algorithms for classifica- tion are both efficient and of high quality, the largest gains can be made from improvements to representation. Document representations are vital for both clustering and classification. Representations exploit the content and structure of documents. Dimensionality reduction can improve the effectiveness of existing representations in terms of quality and run-time performance. Research into these areas is another way to improve the efficiency and quality of clustering and classification results. Evaluating document clustering is a difficult task. Intrinsic measures of quality such as distortion only indicate how well an algorithm minimised a sim- ilarity function in a particular vector space. Intrinsic comparisons are inherently limited by the given representation and are not comparable between different representations. Extrinsic measures of quality compare a clustering solution to a “ground truth” solution. This allows comparison between different approaches. As the “ground truth” is created by humans it can suffer from the fact that not every human interprets a topic in the same manner. Whether a document belongs to a particular topic or not can be subjective.

Experimental warming shows that decomposition temperature sensitivity increases with soil organic matter recalcitrance

Soil C decomposition is sensitive to changes in temperature, and even small increases in temperature may prompt large releases of C from soils. But much of what we know about soil C responses to global change is based on short-term incubation data and model output that implicitly assumes soil C pools are composed of organic matter fractions with uniform temperature sensitivities. In contrast, kinetic theory based on chemical reactions suggests that older, more-resistant C fractions may be more temperature sensitive. Recent research on the subject is inconclusive, indicating that the temperature sensitivity of labile soil organic matter (OM) decomposition could either be greater than, less than, or equivalent to that of resistant soil OM. We incubated soils at constant temperature to deplete them of labile soil OM and then successively assessed the CO2-C efflux in response to warming. We found that the decomposition response to experimental warming early during soil incubation (when more labile C remained) was less than that later when labile C was depleted. These results suggest that the temperature sensitivity of resistant soil OM pools is greater than that for labile soil OM and that global change-driven soil C losses may be greater than previously estimated.

Sensitivity of organic matter decomposition to warming varies with its quality

The relationship between organic matter (OM) lability and temperature sensitivity is disputed, with recent observations suggesting that responses of relatively more resistant OM to increased temperature could be greater than, equivalent to, or less than responses of relatively more labile OM. This lack of clear understanding limits the ability to forecast carbon (C) cycle responses to temperature changes. Here, we derive a novel approach (denoted Q(10-q)) that accounts for changes in OM quality during decomposition and use it to analyze data from three independent sources. Results from new laboratory soil incubations (labile Q(10-q)=2.1 +/- 0.2; more resistant Q(10-q)=3.8 +/- 0.3) and reanalysis of data from other soil incubations reported in the literature (labile Q(10-q)=2.3; more resistant Q(10-q)=3.3) demonstrate that temperature sensitivity of soil OM decomposition increases with decreasing soil OM lability. Analysis of data from a cross-site, field litter bag decomposition study (labile Q(10-q)=3.3 +/- 0.2; resistant Q(10-q)=4.9 +/- 0.2) shows that litter OM follows the same pattern, with greater temperature sensitivity for more resistant litter OM. Furthermore, the initial response of cultivated soils, presumably containing less labile soil OM (Q(10-q)=2.4 +/- 0.3) was greater than that for undisturbed grassland soils (Q(10-q)=1.7 +/- 0.1). Soil C losses estimated using this approach will differ from previous estimates as a function of the magnitude of the temperature increase and the proportion of whole soil OM comprised of compounds sensitive to temperature over that temperature range. It is likely that increased temperature has already prompted release of significant amounts of C to the atmosphere as CO2. Our results indicate that future losses of litter and soil C may be even greater than previously supposed.

Iteratively specified tree-based regression : theory and trip generation example

Focuses on a study which introduced an iterative modeling method that combines properties of ordinary least squares (OLS) with hierarchical tree-based regression (HTBR) in transportation engineering. Information on OLS and HTBR; Comparison and contrasts of OLS and HTBR; Conclusions.

Reliability evaluations of railway power supplies by fault-tree analysis

High reliability of railway power systems is one of the essential criteria to ensure quality and cost-effectiveness of railway services. Evaluation of reliability at system level is essential for not only scheduling maintenance activities, but also identifying reliability-critical components. Various methods to compute reliability on individual components or regularly structured systems have been developed and proven to be effective. However, they are not adequate for evaluating complicated systems with numerous interconnected components, such as railway power systems, and locating the reliability critical components. Fault tree analysis (FTA) integrates the reliability of individual components into the overall system reliability through quantitative evaluation and identifies the critical components by minimum cut sets and sensitivity analysis. The paper presents the reliability evaluation of railway power systems by FTA and investigates the impact of maintenance activities on overall reliability. The applicability of the proposed methods is illustrated by case studies in AC railways.

Applications on fault tree analysis in railway power supply systems

Fault tree analysis (FTA) is presented to model the reliability of a railway traction power system in this paper. First, the construction of fault tree is introduced to integrate components in traction power systems into a fault tree; then the binary decision diagram (BDD) method is used to evaluate fault trees qualitatively and quantitatively. The components contributing to the reliability of overall system are identified with their relative importance through sensitivity analysis. Finally, an AC traction power system is evaluated by the proposed methods.

Clone detection in repositories of business process models

As organizations reach to higher levels of business process management maturity, they often find themselves maintaining repositories of hundreds or even thousands of process models, representing valuable knowledge about their operations. Over time, process model repositories tend to accumulate duplicate fragments (also called clones) as new process models are created or extended by copying and merging fragments from other models. This calls for methods to detect clones in process models, so that these clones can be refactored as separate subprocesses in order to improve maintainability. This paper presents an indexing structure to support the fast detection of clones in large process model repositories. The proposed index is based on a novel combination of a method for process model decomposition (specifically the Refined Process Structure Tree), with established graph canonization and string matching techniques. Experiments show that the algorithm scales to repositories with hundreds of models. The experimental results also show that a significant number of non-trivial clones can be found in process model repositories taken from industrial practice.

Frequency decomposition techniques for increased discriminative 3D facial information capture

Eigen-based techniques and other monolithic approaches to face recognition have long been a cornerstone in the face recognition community due to the high dimensionality of face images. Eigen-face techniques provide minimal reconstruction error and limit high-frequency content while linear discriminant-based techniques (fisher-faces) allow the construction of subspaces which preserve discriminatory information. This paper presents a frequency decomposition approach for improved face recognition performance utilising three well-known techniques: Wavelets; Gabor / Log-Gabor; and the Discrete Cosine Transform. Experimentation illustrates that frequency domain partitioning prior to dimensionality reduction increases the information available for classification and greatly increases face recognition performance for both eigen-face and fisher-face approaches.

Thermal behavior and decomposition of kaolinite-potassium acetate intercalation composite

A series of kaolinite-potassium acetate intercalation composite was prepared. The thermal behavior and decomposition of these composites were investigated by simultaneous differential scanning calorimetry-thermogravimetric analysis (DSC-TGA), X-ray diffraction (XRD) and Fourier-transformation infrared (FT-IR). The XRD pattern at room temperature indicated that intercalation of potassium acetate into kaolinite causes an increase of the basal spacing from 0.718 to 1.428nm. The peak intensity of the expanded phase of the composite decreased with heating above 300°C, and the basal spacing reduced to 1.19nm at 350°C and 0.718nm at 400°C. These were supported by DSC-TGA and FT-IR measurements, where the endothermic reactions are observed between 300 and 600°C. These reactions can be divided into two stages: 1) Removal of the intercalated molecules between 300-400°C. 2) Dehydroxylation of kaolinite between 400-600°C. Significant changes were observed in the infrared bands assigned to outer surface hydroxyl, inner surface hydroxyl, inner hydroxyl and hydrogen bands.

Thermal decomposition of hydrotalcite with hexacyanoferrite(II) and hexacyanoferrate(III) anions in the interlayer : a controlled rate thermal analysis study

The mechanism for the decomposition of hydrotalcite remains unsolved. Controlled rate thermal analysis enables this decomposition pathway to be explored. The thermal decomposition of hydrotalcites with hexacyanoferrite(II) and hexacyanoferrate(III) in the interlayer has been studied using controlled rate thermal analysis technology. X-ray diffraction shows the hydrotalcites studied have a d(003) spacing of 11.1 and 10.9 Å which compares with a d-spacing of 7.9 and 7.98 Å for the hydrotalcite with carbonate or sulphate in the interlayer. Calculations based upon CRTA measurements show that 7 moles of water is lost, proving the formula of hexacyanoferrite(II) intercalated hydrotalcite is Mg6Al2(OH)16[Fe(CN)6]0.5 .7 H2O and for the hexacyanoferrate(III) intercalated hydrotalcite is Mg6Al2(OH)16[Fe(CN)6]0.66 * 9 H2O. Dehydroxylation combined with CN unit loss occurs in three steps between a) 310 and 367°C b) 367 and 390°C and c) between 390 and 428°C for both the hexacyanoferrite(II) and hexacyanoferrate(III) intercalated hydrotalcite.

Thermal stability of the ‘cave’ mineral brushite CaHPO4•2H2O -mechanism of formation and decomposition

Thermogravimetry combined with evolved gas mass spectrometry has been used to ascertain the stability of the ‘cave’ mineral brushite. X-ray diffraction shows that brushite from the Jenolan Caves is very pure. Thermogravimetric analysis coupled with ion current mass spectrometry shows a mass loss at 111°C due to loss of water of hydration. A further decomposition step occurs at 190°C with the conversion of hydrogen phosphate to a mixture of calcium ortho-phosphate and calcium pyrophosphate. TG-DTG shows the mineral is not stable above 111°C. A mechanism for the formation of brushite on calcite surfaces is proposed, and this mechanism has relevance to the formation of brushite in urinary tracts.

Pyktree : a K-tree implementation in Python

In this paper we present pyktree, an implementation of the K-tree algorithm in the Python programming language. The K-tree algorithm provides highly balanced search trees for vector quantization that scales up to very large data sets. Pyktree is highly modular and well suited for rapid-prototyping of novel distance measures and centroid representations. It is easy to install and provides a python package for library use as well as command line tools.