Evolutionary model trees for handling continuous classes in machine learning

Model trees are a particular case of decision trees employed to solve regression problems. They have the advantage of presenting an interpretable output, helping the end-user to get more confidence in the prediction and providing the basis for the end-user to have new insight about the data, confirming or rejecting hypotheses previously formed. Moreover, model trees present an acceptable level of predictive performance in comparison to most techniques used for solving regression problems. Since generating the optimal model tree is an NP-Complete problem, traditional model tree induction algorithms make use of a greedy top-down divide-and-conquer strategy, which may not converge to the global optimal solution. In this paper, we propose a novel algorithm based on the use of the evolutionary algorithms paradigm as an alternate heuristic to generate model trees in order to improve the convergence to globally near-optimal solutions. We call our new approach evolutionary model tree induction (E-Motion). We test its predictive performance using public UCI data sets, and we compare the results to traditional greedy regression/model trees induction algorithms, as well as to other evolutionary approaches. Results show that our method presents a good trade-off between predictive performance and model comprehensibility, which may be crucial in many machine learning applications. (C) 2010 Elsevier Inc. All rights reserved.

Border trees of complex networks

The comprehensive characterization of the structure of complex networks is essential to understand the dynamical processes which guide their evolution. The discovery of the scale-free distribution and the small-world properties of real networks were fundamental to stimulate more realistic models and to understand important dynamical processes related to network growth. However, the properties of the network borders (nodes with degree equal to 1), one of its most fragile parts, remained little investigated and understood. The border nodes may be involved in the evolution of structures such as geographical networks. Here we analyze the border trees of complex networks, which are defined as the subgraphs without cycles connected to the remainder of the network (containing cycles) and terminating into border nodes. In addition to describing an algorithm for identification of such tree subgraphs, we also consider how their topological properties can be quantified in terms of their depth and number of leaves. We investigate the properties of border trees for several theoretical models as well as real-world networks. Among the obtained results, we found that more than half of the nodes of some real-world networks belong to the border trees. A power-law with cut-off was observed for the distribution of the depth and number of leaves of the border trees. An analysis of the local role of the nodes in the border trees was also performed.

Limit theorems for sequences of random trees

We consider a random tree and introduce a metric in the space of trees to define the ""mean tree"" as the tree minimizing the average distance to the random tree. When the resulting metric space is compact we have laws of large numbers and central limit theorems for sequence of independent identically distributed random trees. As application we propose tests to check if two samples of random trees have the same law.

Bootstrap percolation on homogeneous trees has 2 phase transitions

We study the threshold theta bootstrap percolation model on the homogeneous tree with degree b + 1, 2 <= theta <= b, and initial density p. It is known that there exists a nontrivial critical value for p, which we call p(f), such that a) for p > p(f), the final bootstrapped configuration is fully occupied for almost every initial configuration, and b) if p < p(f) , then for almost every initial configuration, the final bootstrapped configuration has density of occupied vertices less than 1. In this paper, we establish the existence of a distinct critical value for p, p(c), such that 0 < p(c) < p(f), with the following properties: 1) if p <= p(c), then for almost every initial configuration there is no infinite cluster of occupied vertices in the final bootstrapped configuration; 2) if p > p(c), then for almost every initial configuration there are infinite clusters of occupied vertices in the final bootstrapped configuration. Moreover, we show that 3) for p < p(c), the distribution of the occupied cluster size in the final bootstrapped configuration has an exponential tail; 4) at p = p(c), the expected occupied cluster size in the final bootstrapped configuration is infinite; 5) the probability of percolation of occupied vertices in the final bootstrapped configuration is continuous on [0, p(f)] and analytic on (p(c), p(f) ), admitting an analytic continuation from the right at p (c) and, only in the case theta = b, also from the left at p(f).

Examples from trees, related to discrete subsets, pseudo-radiality and omega-boundedness

We construct some examples using trees. Some of them are consistent counterexamples for the discrete reflection of certain topological properties. All the properties dealt with here were already known to be non-discretely reflexive if we assume CH and we show that the same is true assuming the existence of a Suslin tree. In some cases we actually get some ZFC results. We construct also, using a Suslin tree, a compact space that is pseudo-radial but it is not discretely generated. With a similar construction, but using an Aronszajn tree, we present a ZFC space that is first countable, omega-bounded but is not strongly w-bounded, answering a question of Peter Nyikos. (C) 2008 Elsevier B.V. All rights reserved.

Nuclear magnetic resonance characterization of metabolite disorder in orange trees caused by citrus sudden death disease

Citrus sudden death (CSD) is a new disease of sweet orange and mandarin trees grafted on Rangpur lime and Citrus volkameriana rootstocks. It was first seen in Brazil in 1999, and has since been detected in more than four million trees. The CSD causal agent is unknown and the current hypothesis involves a virus similar to Citrus tristeza virus or a new virus named Citrus sudden death-associated virus. CSD symptoms include generalized foliar discoloration, defoliation and root death, and, in most cases, it can cause tree death. One of the unique characteristics of CSD disease is the presence of a yellow stain in the rootstock bark near the bud union. This region also undergoes profound anatomical changes. In this study, we analyse the metabolic disorder caused by CSD in the bark of sweet orange grafted on Rangpur lime by nuclear magnetic resonance (NMR) spectroscopy and imaging. The imaging results show the presence of a large amount of non-functional phloem in the rootstock bark of affected plants. The spectroscopic analysis shows a high content of triacylglyceride and sucrose, which may be related to phloem blockage close to the bud union. We also propose that, without knowing the causal CSD agent, the determination of oil content in rootstock bark by low-resolution NMR can be used as a complementary method for CSD diagnosis, screening about 300 samples per hour.

Vedegenskaper i härskande och behärskade granar vid Remingstorp : Wood properties in dominant and sub-dominant trees of Norway spruce at Remingstorp

Variation in wood properties for Picea abies trees and logs of different dimensions has been studied at two sites in southern Sweden of different site quality class. Trees have been classified as dominant or sub-dominant, according to their height. Log and board grades were classified and strength grade of boards, basic density and annual ring width measured. A similar study made on four northern sites was used as reference material.Sub-dominant trees were of superior quality in comparison to dominant trees, when classified by log and board grades or strength grading. Differences were accentuated for the second log where the sub-dominant trees had superior strength and low amount of boards with coarse branches. The results correspond well to those from the northern region, Jämtland. The classifica¬tion of boards as well as bending strength indicated superior properties on timber from northern sites even though the basic density was similar.

Predictive models for chronic renal disease using decision trees, naïve bayes and case-based methods

Data mining can be used in healthcare industry to “mine” clinical data to discover hidden information for intelligent and affective decision making. Discovery of hidden patterns and relationships often goes intact, yet advanced data mining techniques can be helpful as remedy to this scenario. This thesis mainly deals with Intelligent Prediction of Chronic Renal Disease (IPCRD). Data covers blood, urine test, and external symptoms applied to predict chronic renal disease. Data from the database is initially transformed to Weka (3.6) and Chi-Square method is used for features section. After normalizing data, three classifiers were applied and efficiency of output is evaluated. Mainly, three classifiers are analyzed: Decision Tree, Naïve Bayes, K-Nearest Neighbour algorithm. Results show that each technique has its unique strength in realizing the objectives of the defined mining goals. Efficiency of Decision Tree and KNN was almost same but Naïve Bayes proved a comparative edge over others. Further sensitivity and specificity tests are used as statistical measures to examine the performance of a binary classification. Sensitivity (also called recall rate in some fields) measures the proportion of actual positives which are correctly identified while Specificity measures the proportion of negatives which are correctly identified. CRISP-DM methodology is applied to build the mining models. It consists of six major phases: business understanding, data understanding, data preparation, modeling, evaluation, and deployment.

Campus Life: In Perkins Arboretum there are birds of all feathers, trees that tell a story, and even romance (especially if you're a frog)

To the uninitiated, Perkins Arboretum is the edge of campus. For those familiar with its flora and fauna, the arboretum is a wondrous classroom and refuge.

Occurrence of Quesada gigas on Schizolobium amazonicum trees in Maranhão and Pará States, Brazil

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Hepatic concentration of heat shock protein 70 kD (Hsp70) in broilers subjected to different thermal treatments

1. The relationship between repeated thermal treatments and hepatic synthesis of Hsp 70 was studied in broiler chickens.2. Sixty broilers were submitted to 5 different treatments (12 birds each) from day 1 to day 42. Four groups were kept in a thermoneutral environment and subjected to 0, 1, 2 and 3 heat stress episodes at 35 degrees C for 4 h per week (TN-0, TN-1, TN-2 and TN-3, respectively). The last group (HT-35) was reared at a room temperature of 35 degrees C.3. From 39 to 42 old, the birds experienced acute heat stress at 41 degrees C. Resistance to heat stress was evaluated by the time taken for rectal temperature to increase by 3 degrees C above the pre-treatment value. Livers were collected (before and after heat stress) and Hsp70 was determined using Western Blot analysis with monoclonal anti-Hsp70 antibody.4. Resistance to heat stress and concentration of Hsp70 were higher in those birds subjected to more heat stress episodes during the experimental period (TN-3) and HT-35. A positive correlation was observed between Hsp70 concentration and the time taken for a 3 degrees C increase in rectal temperature (r=0.42; P<0.01).5. Exposing birds to episodes of heat stress (35 degrees C) during rearing may improve their resistance to acute heat stress, but the previous thermal history did not seem to influence the hepatocyte Hsp70 content after exposure to more severe heat stress (41 degrees C).