Kernel-based machine learning for fast text mining in R

Recent advances in the field of kernel-based machine learning methods allow fast processing of text using string kernels utilizing suffix arrays. kernlab provides both kernel methods’ infrastructure and a large collection of already implemented algorithms and includes an implementation of suffix-array-based string kernels. Along with the use of the text mining infrastructure provided by tm these packages provide R with functionality in processing, visualizing and grouping large collections of text data using kernel methods. The emphasis is on the performance of various types of string kernels at these tasks.     

A Simulation Framework for Heterogeneous Agents

We introduce a generic simulation framework suitable for agent-based simulations featuring the support of heterogeneous agents, hierarchical scheduling, and flexible specification of design parameters. One key aspect of this framework is the design specification: we use a format based on the Extendible Markup Language (XML) that is simple-structured yet still enables the design of flexible models. Another issue in agent-based simulations, especially when ready-made components are used, is the heterogeneity arising from both the agents' implementations and the underlying platforms. To tackle such obstacles, we introduce a wrapper technique for mapping the functionality of agents living in an interpreter-based environment to a standardized JAVA interface, thus facilitating the task for any control mechanism (like a simulation manager) because it has to handle only one set of commands for all agents involved. Again, this mapping is made by an XML-based definition format. We demonstrate the technique by applying it to a simple sample simulation of two mass marketing firms operating in an artificial consumer environment.     

Effect of minerals on casein micelle stability of cows' milk

The effects of minerals on casein micelle stability of individual cows' milk, throughout a complete lactation, were investigated. Calcium and calcium ions, magnesium, phosphorus, sodium, potassium and citrate contents were analysed, together with the following physical properties of milk; pH, ethanol stability, rennet clotting time and coagulum firmness. There was an inverse non-linear relationship between free calcium ion concentration and ethanol stability (ES; r=0.84). Rennet coagulation time showed a weaker relationship with free calcium ion concentration (r=0.44) but a stronger relationship with pH (r=0.66). In addition, samples containing higher amounts of free calcium ions produced a firmer gel. Citrate in natural samples acts as a stabilizing factor, as it slightly improves milk stability. Potassium, on the other hand, exhibited a negative correlation, but only with rennet clotting time (r=-0.52). Throughout lactation the average values were; free Ca2+ concentration 1.88 mM, pH 6.63, ES 83.2% and clotting time 13.6 min. The equilibrium relationship between pH and free Ca2+ concentration was investigated by adjusting milk pH from 5.9 to 7.1, using acid and alkali. There was a good inverse linear relationship between pH and log (free Ca2+) for individual milk samples, with a gradient of -0.62 and a standard deviation of 0.042.     

Quantifiable data mining using ratio rules

Association Rule Mining algorithms operate on a data matrix (e.g., customers products) to derive association rules [AIS93b, SA96]. We propose a new paradigm, namely, Ratio Rules, which are quantifiable in that we can measure the “goodness” of a set of discovered rules. We also propose the “guessing error” as a measure of the “goodness”, that is, the root-mean-square error of the reconstructed values of the cells of the given matrix, when we pretend that they are unknown. Another contribution is a novel method to guess missing/hidden values from the Ratio Rules that our method derives. For example, if somebody bought $10 of milk and $3 of bread, our rules can “guess” the amount spent on butter. Thus, unlike association rules, Ratio Rules can perform a variety of important tasks such as forecasting, answering “what-if” scenarios, detecting outliers, and visualizing the data. Moreover, we show that we can compute Ratio Rules in a single pass over the data set with small memory requirements (a few small matrices), in contrast to association rule mining methods which require multiple passes and/or large memory. Experiments on several real data sets (e.g., basketball and baseball statistics, biological data) demonstrate that the proposed method: (a) leads to rules that make sense; (b) can find large itemsets in binary matrices, even in the presence of noise; and (c) consistently achieves a “guessing error” of up to 5 times less than using straightforward column averages. Received: March 15, 1999 / Accepted: November 1, 1999     

Effects of carbonate on the electrolytic removal of ammonia and urea from urine with thermally prepared IrO<Subscript>2</Subscript> electrodes

Recent studies have shown that electrolysis can be an efficient process for nitrogen removal from urine. These studies have been conducted with urea solutions or fresh urine, but urine collected in NoMix toilets and urinals has a substantially different composition, because bacteria hydrolyse urea quickly to ammonia and carbonate. In this study, we compared electrochemical removal of nitrogen from synthetic solutions of fresh and stored urine using IrO₂ anodes. We could show that in fresh urine both ammonia and urea are efficiently eliminated, mainly through chlorine-mediated oxidation. However, in stored urine the presence of carbonate, arising from urea hydrolysis, leads to an inhibition of ammonia oxidation. We suggest two parallel mechanisms to explain this effect: the competition between chloride and carbonate oxidation at the anode and the competition between chlorate formation, enhanced by the buffering effect of carbonate, and ammonia oxidation for the consumption of active chlorine in the bulk. However, further experiments are needed to support the latter mechanism. In conclusion, this study highlights the negative consequences of the presence of carbonate in urine solutions, but also in other wastewaters, when subjected to an electrolytic treatment on IrO₂ in alkaline media.     

Multi-view human movement recognition based on fuzzy distances and linear discriminant analysis

In this paper, a novel multi-view human movement recognition method is presented. A novel representation of multi-view human movement videos is proposed that is based on learning basic multi-view human movement primitives, called multi-view dynemes. The movement video is represented in a new feature space (called dyneme space) using these multi-view dynemes, thus producing a time invariant multi-view movement representation. Fuzzy distances from the multi-view dynemes are used to represent the human body postures in the dyneme space. Three variants of Linear Discriminant Analysis (LDA) are evaluated to achieve a discriminant movement representation in a low dimensionality space. The view identification problem is solved either by using a circular block shift procedure followed by the evaluation of the minimum Euclidean distance from any dyneme, or by exploiting the circular shift invariance property of the Discrete Fourier Transform (DFT). The discriminant movement representation combined with camera viewpoint identification and a nearest centroid classification step leads to a high human movement classification accuracy.     

Thymic Epithelial Neoplasms: Focusing on the Epigenetic Alterations

Thymic Epithelial Neoplasms (TENs) represent the most common tumors of the thymus gland. Epigenetic alterations are generally involved in initiation and progression of various cancer entities. However, little is known about the role of epigenetic modifications in TENs. In order to identify relevant studies, a literature review was conducted using the MEDLINE and LIVIVO databases. The search terms thymoma, thymic carcinoma, thymic epithelial neoplasm, epigenetics, DNA methylation, HDAC and miRNA were employed and we were able to identify forty studies focused on TENs and published between 1997 and 2021. Aberrant epigenetic alterations seem to be involved in the tumorigenesis of thymomas and thymic carcinomas, with numerous studies reporting on non-coding RNA clusters and altered gene methylation as possible biomarkers in different types of TENs. Interestingly, Histone Deacetylase Inhibitors have shown potent antitumor effects in clinical trials, thus possibly representing effective epigenetic therapeutic agents in TENs. Additional studies in larger patient cohorts are, nevertheless, needed to verify the clinical utility and safety of novel epigenetic agents in the treatment of patients with TENs.