PMCRI: a parallel modular classification rule induction framework

In a world where massive amounts of data are recorded on a large scale we need data mining technologies to gain knowledge from the data in a reasonable time. The Top Down Induction of Decision Trees (TDIDT) algorithm is a very widely used technology to predict the classification of newly recorded data. However alternative technologies have been derived that often produce better rules but do not scale well on large datasets. Such an alternative to TDIDT is the PrismTCS algorithm. PrismTCS performs particularly well on noisy data but does not scale well on large datasets. In this paper we introduce Prism and investigate its scaling behaviour. We describe how we improved the scalability of the serial version of Prism and investigate its limitations. We then describe our work to overcome these limitations by developing a framework to parallelise algorithms of the Prism family and similar algorithms. We also present the scale up results of a first prototype implementation.

Parallel induction of modular classification rules

The Distributed Rule Induction (DRI) project at the University of Portsmouth is concerned with distributed data mining algorithms for automatically generating rules of all kinds. In this paper we present a system architecture and its implementation for inducing modular classification rules in parallel in a local area network using a distributed blackboard system. We present initial results of a prototype implementation based on the Prism algorithm.

Parallel rule induction with information theoretic pre-pruning

In a world where data is captured on a large scale the major challenge for data mining algorithms is to be able to scale up to large datasets. There are two main approaches to inducing classification rules, one is the divide and conquer approach, also known as the top down induction of decision trees; the other approach is called the separate and conquer approach. A considerable amount of work has been done on scaling up the divide and conquer approach. However, very little work has been conducted on scaling up the separate and conquer approach.In this work we describe a parallel framework that allows the parallelisation of a certain family of separate and conquer algorithms, the Prism family. Parallelisation helps the Prism family of algorithms to harvest additional computer resources in a network of computers in order to make the induction of classification rules scale better on large datasets. Our framework also incorporates a pre-pruning facility for parallel Prism algorithms.

P-Prism: a computationally efficient approach to scaling up classification rule induction

Top Down Induction of Decision Trees (TDIDT) is the most commonly used method of constructing a model from a dataset in the form of classification rules to classify previously unseen data. Alternative algorithms have been developed such as the Prism algorithm. Prism constructs modular rules which produce qualitatively better rules than rules induced by TDIDT. However, along with the increasing size of databases, many existing rule learning algorithms have proved to be computational expensive on large datasets. To tackle the problem of scalability, parallel classification rule induction algorithms have been introduced. As TDIDT is the most popular classifier, even though there are strongly competitive alternative algorithms, most parallel approaches to inducing classification rules are based on TDIDT. In this paper we describe work on a distributed classifier that induces classification rules in a parallel manner based on Prism.

Towards a computationally efficient approach to modular classification rule induction

Induction of classification rules is one of the most important technologies in data mining. Most of the work in this field has concentrated on the Top Down Induction of Decision Trees (TDIDT) approach. However, alternative approaches have been developed such as the Prism algorithm for inducing modular rules. Prism often produces qualitatively better rules than TDIDT but suffers from higher computational requirements. We investigate approaches that have been developed to minimize the computational requirements of TDIDT, in order to find analogous approaches that could reduce the computational requirements of Prism.

Induction of modular classification rules: using Jmax-pruning

The Prism family of algorithms induces modular classification rules which, in contrast to decision tree induction algorithms, do not necessarily fit together into a decision tree structure. Classifiers induced by Prism algorithms achieve a comparable accuracy compared with decision trees and in some cases even outperform decision trees. Both kinds of algorithms tend to overfit on large and noisy datasets and this has led to the development of pruning methods. Pruning methods use various metrics to truncate decision trees or to eliminate whole rules or single rule terms from a Prism rule set. For decision trees many pre-pruning and postpruning methods exist, however for Prism algorithms only one pre-pruning method has been developed, J-pruning. Recent work with Prism algorithms examined J-pruning in the context of very large datasets and found that the current method does not use its full potential. This paper revisits the J-pruning method for the Prism family of algorithms and develops a new pruning method Jmax-pruning, discusses it in theoretical terms and evaluates it empirically.

Scaling up classification rule induction through parallel processing

The fast increase in the size and number of databases demands data mining approaches that are scalable to large amounts of data. This has led to the exploration of parallel computing technologies in order to perform data mining tasks concurrently using several processors. Parallelization seems to be a natural and cost-effective way to scale up data mining technologies. One of the most important of these data mining technologies is the classification of newly recorded data. This paper surveys advances in parallelization in the field of classification rule induction.

Computationally efficient induction of classification rules with the PMCRI and J-PMCRI frameworks

In order to gain knowledge from large databases, scalable data mining technologies are needed. Data are captured on a large scale and thus databases are increasing at a fast pace. This leads to the utilisation of parallel computing technologies in order to cope with large amounts of data. In the area of classification rule induction, parallelisation of classification rules has focused on the divide and conquer approach, also known as the Top Down Induction of Decision Trees (TDIDT). An alternative approach to classification rule induction is separate and conquer which has only recently been in the focus of parallelisation. This work introduces and evaluates empirically a framework for the parallel induction of classification rules, generated by members of the Prism family of algorithms. All members of the Prism family of algorithms follow the separate and conquer approach.

Carbon monoxide induces cardiac arrhythmia via induction of the late Na+ current

Our data indicate that the proarrhythmic effects of CO arise from activation of NO synthase, leading to NO-mediated nitrosylation of Na(V)1.5 and to induction of the late Na(+) current. We also show that the antianginal drug ranolazine can abolish CO-induced early after-depolarizations, highlighting a novel approach to the treatment of CO-induced arrhythmias.

The induction of mental and contact contamination

Background: Extreme fear of contamination within Obsessive Compulsive Disorder is traditionally conceptualised as a physical phenomenon. More recent research has supported the notion of ‘mental’ contamination, in which people feel contaminated in the absence of physical contact. The current research sought to determine whether feelings of contact and mental contamination could be induced within a non-clinical sample, whether the impact of mental and contact contamination was comparable in terms of associated feelings and behaviour and whether related psychopathology related to the impact of the tasks. Methods: Undergraduate students (n=60) completed OCD relevant measures and were randomly assigned to either a contact contamination condition (CC: moving a bucket of fake vomit) or a mental contamination condition (MC: thinking about a bucket of vomit). Results: Both manipulations induced feelings of contamination. Participants in the contact condition had significantly greater urges to wash than those in the mental condition. Neutralising behaviour did not differ across conditions. Conclusions: Feelings of contamination can be induced in the absence of physical contact and for those in the MC group, some aspects of OCD-relevant psychopathology were related to the impact of the manipulation. These findings have implications for the understanding and treatment of contamination-related fears in OCD.

Repressor element 1 silencing transcription factor couples loss of pluripotency with neural induction and neural differentiation

Neural differentiation of embryonic stem cells (ESCs) requires coordinated repression of the pluripotency regulatory program and reciprocal activation of the neurogenic regulatory program. Upon neural induction, ESCs rapidly repress expression of pluripotency genes followed by staged activation of neural progenitor and differentiated neuronal and glial genes. The transcriptional factors that underlie maintenance of pluripotency are partially characterized whereas those underlying neural induction are much less explored, and the factors that coordinate these two developmental programs are completely unknown. One transcription factor, REST (repressor element 1 silencing transcription factor), has been linked with terminal differentiation of neural progenitors and more recently, and controversially, with control of pluripotency. Here, we show that in the absence of REST, coordination of pluripotency and neural induction is lost and there is a resultant delay in repression of pluripotency genes and a precocious activation of both neural progenitor and differentiated neuronal and glial genes. Furthermore, we show that REST is not required for production of radial glia-like progenitors but is required for their subsequent maintenance and differentiation into neurons, oligodendrocytes, and astrocytes. We propose that REST acts as a regulatory hub that coordinates timely repression of pluripotency with neural induction and neural differentiation.

The effect of proteolysis on the induction of cell death by monomeric alpha-lactalbumin.

α-Lactalbumin (α-la) is a major whey protein found in milk. Previous data suggested that α-la has antiproliferative effects in human adenocarcinoma cell lines such as Caco-2 and HT-29. However, the cell death inducing α-la was not a naturally occurring monomer but either a multimeric variant or an α-la:oleic acid complex (HAMLET/BAMLET). Proteolysis showed that both human and bovine α-la are susceptible to digestion. ELISA assays assessing cell death with the native undigested α-la fractions showed that undigested protein fractions did have a significant cell death effect on CaCo-2 cells. Bovine α-la was also more effective than human α-la. A reduction in activity corresponded with lower concentrations of the protein and partial digestion and fragmentation of the protein using trypsin and pepsin. This suggests that the tertiary structure is vital for the apoptotic effect.

Induction of cold active acid phosphomonoesterase activity at low temperature in psychrotrophic ectomycorrhizal Hebeloma spp.

Hebeloma strains of arctic and temperature origin, grown at 22° or 6°, were assayed for wall-bound and extracellular acid phosphomonoesterase (pNPPase) across a temperature range 2-37°. Only when grown at 6° was a cold active extracellular pNPPase induced in all the arctic strains and most of the temperature strains tested. Such enzymes are suggested to be a adaptation to low soil temperatures, and are discussed in the context of ectomycorrhizal access to soil PO4− monoesters at low temperature.

Towards expressive rule induction on IP network event streams

In order to gain insights into events and issues that may cause errors and outages in parts of IP networks, intelligent methods that capture and express causal relationships online (in real-time) are needed. Whereas generalised rule induction has been explored for non-streaming data applications, its application and adaptation on streaming data is mostly undeveloped or based on periodic and ad-hoc training with batch algorithms. Some association rule mining approaches for streaming data do exist, however, they can only express binary causal relationships. This paper presents the ongoing work on Online Generalised Rule Induction (OGRI) in order to create expressive and adaptive rule sets real-time that can be applied to a broad range of applications, including network telemetry data streams.

Media damage following detergent sclerotherapy appears to be secondary to the induction of inflammation and apoptosis: an immunohistochemical study elucidating previous histological observations

Objective/Background: Traditionally, sclerotherapy has been thought to work by the cytotoxic effect of the sclerosant upon the endothelium alone. However, studies have shown that sclerotherapy is more successful in smaller veins than in larger veins. This could be explained by the penetration of the sclerosant, or its effect, into the media. This study aimed to investigate intimal and medial damage profiles after sclerosant treatment. Methods: Fresh human varicose veins were treated ex vivo with either 1% or 3% sodium tetradecyl sulphate (STS) for 1 or 10 minutes. The effect of the sclerosant on the vein wall was investigated by immunofluorescent labelling of transverse vein sections using markers for endothelium (CD31), smooth muscle (a-actin), apoptosis (p53) and inflammation (intercellular adhesion molecule-1 [ICAM-1]). Polidocanol (POL; 3%) treatment at 10 minutes was similarly investigated. Results: Endothelial cell death was concentration- and time-dependent for STS but incomplete for both sclerosants. Time, but not concentration, significantly affected cell death (p > .001). A 40% and 30% maximum reduction was observed for STS and POL, respectively. Destruction of 20e30% of smooth muscle cells was found up to 250 mm from the lumen after 3% STS treatment for 10 minutes. POL treatment for 10 minutes showed inferior destruction of medial cells. Following STS treatment and 24-hour tissue culture, p53 and ICAM-1 were upregulated to a depth of around 300 mm. This effect was not observed with POL. Conclusion: Inflammatory and apoptotic markers show the same distribution as medial cell death, implying that sclerotherapy with STS works by inducing apoptosis in the vein wall rather than having an effect restricted to the endothelium. Incomplete loss of endothelial cells and penetration of the sclerosant effect up to 250 mm into the media suggest that medial damage is crucial to the success of sclerotherapy and may explain why it is less effective in larger veins.