Subcellular localisation of proteins in fluorescent microscope images using a random forest

This paper presents a system that employs random forests to formulate a method for subcellular localisation of proteins. A random forest is an ensemble learner that grows classification trees. Each tree produces a classification decision, and an integrated output is calculated. The system classifies the protein-localisation patterns within fluorescent microscope images. 2D images of HeLa cells that include all major classes of subcellular structures, and the associated feature set are used. The performance of the developed system is compared against that of the support vector machine and decision tree approaches. Three experiments are performed to study the influence of the training and test set size on the performance of the examined methods. The calculated classification errors and execution times are presented and discussed. The lowest classification error (2.9%) has been produced by the developed system.

A random forest for lung nodule identification

A method is presented for identification of lung nodules. It includes three stages: image acquisition, background removal, and nodule detection. The first stage improves image quality. The second stage extracts long lobe regions. The third stage detects lung nodules. The method is based on the random forest learner. Training set contains nodule, non-nodule, and false-positive patterns. Test set contains randomly selected images. The developed method is compared against the support vector machine. True-positives of 100% and 85.9%, and false-positives of 1.27 and 1.33 per image were achieved by the developed method and the support vector machine, respectively.

Random walk versus multiple trend breaks in stock prices : evidence from 15 European markets

This letter extends research reported in Narayan and Smyth (2005) by employing multiple trend break unit root tests to examine the random walk hypothesis for 15 European stock market indices. The results provide strong support for the view that stock prices are characterized by a random walk.

Mean reversion versus random walk in G7 stock prices evidence from multiple trend break unit root tests

This paper provides evidence on the random walk hypothesis in G7 stock price indices using unit root tests which allow for one and two structural breaks in the trend. Of the seven countries we find, at best, evidence of mean reversion in the stock price index of Japan. Thus, overall, our results support the random walk hypothesis. We also consider the implications of the identified structural breaks for movement in stock prices over time. Our main conclusion from this exercise is that the second break in stock prices has had a detrimental effect on movements in stock prices in the G7 countries.

A comparison of the ICI Pillbox and the Random Tumble Methods is assessing pilling and appearance change of worsted spun cashmere and cashmere-wool blend knitwear

This study evaluated the differences between two international test methods on the assessment of pilling and appearance change of worsted spun cashmere and superfine wool knitwear and their blends. Differences between the standard ICI Pill Box Method and the Random Tumble Method were found in both the significance and magnitude of resistance to pilling and appearance change and the amount of fabric mass loss of worsted spun cashmere and cashmere superfine wool blend knit fabrics. The ICI Pill Box Method differentiated to a greater extent the effects of wool type and blend ratio of cashmere and wool compared with the Random Tumble Method. Generally the addition of cashmere or low crimp superfine wool resulted in fabrics being more resistance to pilling and appearance change compared with fabrics made from high crimp superfine wool. This was associated with increased fabric mass loss when assessed by the ICI Pill Box Method but not with the Random Tumble Method. KEYWORDS: Cashmere, crimp, wool, pilling, appearance change, knitwear

Testing for a random walk in Italy's inflation in the post-war period, 1947-1996

This paper investigates the unit root properties of Italy’s inflation rate in the post-war period (1947-1996). To achieve the aim of this study, the Zivot and Andrews (1992) one break test and the Lumsdaine and Papell (1997) two breaks test for unit roots are applied. It is found that inflation for Italy was a non-stationary breakpoint for the period 1947-1996. This result has important implications for econometric modeling and in understanding the behavior of shocks to Italy’s inflation.

Testing for the random walk hypothesis in the case of visitor arrivals : evidence from Indian tourism

Testing for the random walk hypothesis, which asserts that a series is a non-stationary process or a unit root process, in the case of visitor arrivals has important implications for policy. If, for instance, visitor arrivals are characterized by a unit root, then it implies that shocks to visitor arrivals are permanent. However, if visitor arrivals are without a unit root, this implies that shocks to visitor arrivals are temporary. This study provides evidence on the random walk hypothesis for visitor arrivals to India using the recently developed Im et al. (2003) and Maddala and Wu (1999) panel unit root tests. Both tests allow one to reject the random walk hypothesis, implying that shocks to visitor arrivals to India from the 10 major source markets have a temporary effect on visitor arrivals.

Are OECD stock prices characterized by a random walk? Evidence from sequential trend break and panel data models

This paper examines whether stock prices for a sample of 22 OECD countries can be best represented as mean reversion or random walk processes. A sequential trend break test proposed by Zivot and Andrews is implemented, which has the advantage that it can take account of a structural break in the series, as well as panel data unit root tests proposed by Im et al., which exploits the extra power in the panel properties of the data. Results provide strong support for the random walk hypothesis.

Is extrapair mating random? On the probability distribution of extrapair young in avian broods

A dichotomy in female extrapair copulation (EPC) behavior, with some females seeking EPC and others not, is inferred if the observed distribution of extrapair young (EPY) over broods differs from a random process on the level of individual offspring (binomial, hypergeometrical, or Poisson). A review of the literature shows such null models are virtually always rejected, with often large effect sizes. We formulate an alternative null model, which assumes that 1) the number of EPC has a random (Poisson) distribution across females (broods) and that 2) the probability for an offspring to be of extrapair origin is zero without any EPC and increases with the number of EPC. Our brood-level model can accommodate the bimodality of both zero and medium rates of EPY typically found in empirical data, and fitting our model to EPY production of 7 passerine bird species shows evidence of a nonrandom distribution of EPY in only 2 species. We therefore argue that 1) dichotomy in extrapair mate choice cannot be inferred only from a significant deviation in the observed distribution of EPY from a random process on the level of offspring and that 2) additional empirical work on testing the contrasting critical predictions from the classic and our alternative null models is required.

Intersection and partition graphs

This thesis reviews previous work done on both general partition graphs and existential partition graphs, which are a generalised form of general partition graphs, and extends some of the results.

Multilabel classification by BCH code and random forests

This paper uses error correcting codes for multilabel classification. BCH code and random forests learner are used to form the proposed method. Thus, the advantage of the error-correcting properties of BCH is merged with the good performance of the random forests learner to enhance the multilabel classification results. Three experiments are conducted on three common benchmark datasets. The results are compared against those of several exiting approaches. The proposed method does well against its counterparts for the three datasets of varying characteristics.

Dual-random ensemble method for multi-label classification of biological data

This paper presents a dual-random ensemble multi-label classification method for classification of multi-label data. The method is formed by integrating and extending the concepts of feature subspace method and random k-label set ensemble multi-label classification method. Experiemental results show that the developed method outperforms the exisiting multi-lable classification methods on three different multi-lable datasets including the biological yeast and genbase datasets.

Random forest based lung nodule classification aided by clustering

An automated lung nodule detection system can help spot lung abnormalities in CT lung images. Lung nodule detection can be achieved using template-based, segmentation-based, and classification-based methods. The existing systems that include a classification component in their structures have demonstrated better performances than their counterparts. Ensemble learners combine decisions of multiple classifiers to form an integrated output. To improve the performance of automated lung nodule detection, an ensemble classification aided by clustering (CAC) method is proposed. The method takes advantage of the random forest algorithm and offers a structure for a hybrid random forest based lung nodule classification aided by clustering. Several experiments are carried out involving the proposed method as well as two other existing methods. The parameters of the classifiers are varied to identify the best performing classifiers. The experiments are conducted using lung scans of 32 patients including 5721 images within which nodule locations are marked by expert radiologists. Overall, the best sensitivity of 98.33% and specificity of 97.11% have been recorded for proposed system. Also, a high receiver operating characteristic (ROC) A_z of 0.9786 has been achieved.