Set Of Decision Trees Research Articles

A System Intervention to Increase Pneumococcal Vaccination Rates in Adults Living With Human Immunodeficiency Virus Who Present for Scheduled Medical Care

Abstract Pneumococcal vaccinations are recommended for adults living with HIV; however, eligibility criteria are complex. At a clinic specializing in the care of lesbian, gay, bisexual, transgender, or queer/questioning persons, only 36% of patients who were vaccine-eligible received the vaccination. A set of decision trees was developed to guide clinic personnel in determining whether a vaccine is recommended. The provider then determined whether the vaccine was appropriate based on each patient's complete clinical presentation. When appropriate, patients were offered the vaccine at the same appointment. This practice change led to a 22% increased vaccination rate.

Decision Rules Induced From Sets of Decision Trees

Decision rules belong to known forms of knowledge representation. Among popular measures of their quality length and support can be distinguished. Shorter rules are easier to understand and interpret. Support allows to present patterns hidden in the data. Nowadays, data mining tasks are oriented toward extracting knowledge from data in both distributed and centralized forms. Learning decision rules from a decision tree is a relatively simple task. However, the challenge arises when decision rules are induced from a set of decision trees. Moreover, in the case of distributed data, the decision trees may be constructed independently on different sources, and merging them into a unified set requires resolving conflicts and inconsistencies. In this paper, decision rules are constructed from distributed data based on decision trees induced using the randomly chosen attributes as the splitting criterion. The aim of the study is to compare the quality of two algorithms for constructing rules which are true for a maximum number of trees. The comparison was made based on three factors: the number of trees for which the rule is true, their length and support. Based on performed experiments it was possible to see that the number of true rules for the maximum number of decision trees from the set is greater for algorithm A than for heuristics H. This algorithm allows the induction of shorter rules with greater support compared to heuristic H. However, it should be also noted that the rules induced by heuristic H are often true for a larger number of trees than the rules constructed by algorithm A. Thus, both algorithms can be applied to distributed data.

Exploring Life Quality Index via Education Level and Population Density

The subjective nature of quality of life, makes it complex to measure the quality of life. Among them, finding a few key criteria that affect the quality of life in cities will be the main purpose. In this study, a linear regression model was used to predict the numerical relationship between population density, bachelor's degree rate and education Gini index, and further Akaike information criterion (AIC) and Bayesian information criterion (BIC) were selected to validate and adjust the model, and finally F-test was used to compare the initial model with the adjusted model. In addition, the random forest algorithm was used to classify the quality of life of each city by population density as well as education level to identify the common characteristics of cities with high physical quality of life index. Three classification decision trees created using different combinations of population density, bachelor's degree rate and education Gini index were used, followed by additional multiple decision trees generated by the random forest algorithm. The final results of the receiver operator characteristic curve (ROC) curves and confusion matrix of the two sets of decision trees were observed and evaluated to stratify the cities. The experimental results show that population density has a negative effect on quality of life, while bachelor's degree rate has a positive effect. Although the three indicators of population density, bachelor's degree rate and education Gini index cannot effectively distinguish the four levels, they can effectively distinguish which cities are not the first level. it shows that cities that achieve high education level and low population density are not necessarily the cities with the highest quality of life, but those that do not are definitely not.

A Novel ML Augmented DRC Framework for Identification of Yield Detractor Patterns

This work proposes a methodology to find lithography yield detractors using Design Rule Checks (DRC) that are derived from a supervised Machine Learning (ML) model. The probability of being an outlier in layout parameter domain has a strong correlation with the probability of process failure. Moreover, the failing patterns exhibit relatively complex and non-linear behavior and often form complex clusters in the layout parameter domain. Using this, an accurate failure model is built by measuring the distance of a layout sample from the mean distribution in the layout parameter space. The proposed method does not require process failure models, but the calculation of layout parameters only. Further, the failure models are converted into DRC rules to make the methodology suitable for integration into present verification flow. These ML augmented Design Rule Checks (MLDRC) use a set of decision trees in the layout parameter domain and are suitable for full-chip level applications. The ML augmented DRC can better represent and form failure clusters as compared to traditional DRC. Experimental results show that the proposed MLDRC achieves better performance on full-chip designs compared to other hotspot detection techniques.

Machine learning imputation of metastatic status from open claims in melanoma patients.

e21540 Background: Metastatic status is a crucial variable in most oncology studies but is not available in claims data. The objective of this study is to develop a machine learning model for Imputation of metastatic status from claims data with ground. Truth is derived from highly curated electronic medical record data. Methods: We used a set of 11389 melanoma patients from the ConcertAI real world database of intersecting claims and EMR data that includes data from CancerLinQ Discovery. Using features from claims and our gold standard labels from EMR we built an ML model using (XGBoost) extreme gradient boosting, an algorithm that iteratively combines a set of decision trees into a single model. We used 60% of the data for training, 20% for hyper-parameter tuning, and 20% for holdout testing. The model was built using 55 features. Results: The table below summarizes results. Metrics are on the final hold out set which was unseen by the model and entirely composed of highly curated EMR data. Conclusions: We are able to build a high precision model for the imputation of metastatic melanoma status using claims data. This could enable significantly better use of claims data stemming from the ability to find a metastatic cohort with very few false positives. Providing more precise cohort identification for comparative effectiveness studies. We found features such as secondary neoplasm diagnosis, anti-neoplastic meds, and radiation ranking highly in our analysis of model feature importances. Using techniques to analyze non-linear feature interactions in our AI model we found an interaction relationship between long term anti-neoplastic therapy, reported pain and metastatic status which we plan to further study. This work is preliminary and we are working to further improve model performance.[Table: see text]

Lameness prediction in broiler chicken using a machine learning technique

Broiler flock welfare is usually assessed through mortality, physiology, behavior, and walking ability. The possibility of assessing broiler chicken lameness using the bird walking ability was investigated using the machine learning approach for the first time. Data on broiler walking speed and acceleration, genetic strain, and sex were recorded and input in a dataset. Broilers were classified according to the 6-point gait score (GS0 is a sound bird, and GS5 is a severely lame bird). Decision trees were built initially using all datasets. The confusion matrix of each developed model was analyzed. The pruning technique was used, removing from the dataset the variables that did not infer in the classification results. We reorganized the dataset and re-arranged the data by grouping the intermediate target class of gait score using the Borda Count method. Re-processing data, we obtained a new set of decision trees. Using the 3-point gait score (GS0 is a sound bird, and GS2 is a lame bird), we obtained a new model with better accuracy (78%); however, the model had a lower accuracy for classifying lame broilers (GS2, 5%). The final decision tree was selected for classifying broilers, either sound or lame, according to their walking speed. The developed model presented good accuracy (91%), and it ordered properly sound (86%) and lame birds (92%). The novel model might be used to assess broiler lameness on-farm by registering the bird displacement velocity. Further developments using the model might allow flock lameness detection automatically.

Embarcadero: Species distribution modelling with Bayesian additive regression trees in r

Abstract embarcadero is an r package of convenience tools for species distribution modelling (SDM) with Bayesian additive regression trees (BART), a powerful machine learning approach that has been rarely applied to ecological problems. Like other classification and regression tree methods, BART estimates the probability of a binary outcome based on a set of decision trees. Unlike other methods, BART iteratively generates sets of trees based on a set of priors about tree structure and nodes, and builds a posterior distribution of estimated classification probabilities. So far, BARTs have yet to be applied to SDM. embarcadero is a workflow wrapper for BART species distribution models, and includes functionality for easy spartial prediction, an automated variable selection procedure, several types of partial dependence visualization and other tools for ecological application. The embarcadero package is an open source and available on Github. To show how embarcadero can be used by ecologists, I illustrate a BART workflow for a virtual species distribution model. The supplement includes a more advanced vignette showing how BART can be used for mapping disease transmission risk, using the example of Crimean–Congo haemorrhagic fever in Africa.

ALGORITHMS FOR MANAGING THE LOGICAL STRUCTURE OF THE DATABASE USING A PARAMETRIC MODEL OF COMPETITIVE ACCESS TO QUERIES BASED ON THE RANDOM FOREST METHOD

In article discusses the approach to development of mathematical software for support the process of managing the data schema in relational DBMS in terms of processing of parallel queries stream that compete for data in the hierarchy of the DBMS core memory. The necessity of the formation of a parametric model of queries competitive access. Briefly discusses methods of machine learning, allowing to solve the problem of regression recovery. The use of the random forest method as the most universal method of approximation of arbitrary functions is substantiated. A method of forming a parametric model of competitive access based on the random forest method, as well as an approach with the ensemble of sets of decision trees, which allows to provide the required generalizing ability and stability of the model to partial features and diversity of all types of queries received at the input of the DBMS. The stages of the developed algorithms are presented: ranking query parameters by total execution time and automatic data distribution, allowing you to go from approximating the target system with linear-continuous functions to a set of logical data schema objects, ordered by their effect on time, total query execution time, reducing multi-criteria optimization task to a task optimization by one criterion.

Formal Verification of a Decision-Tree Ensemble Model and Detection of Its Violation Ranges

As one type of machine-learning model, a "decision-tree ensemble model" (DTEM) is represented by a set of decision trees. A DTEM is mainly known to be valid for structured data; however, like other machine-learning models, it is difficult to train so that it returns the correct output value for any input value. Accordingly, when a DTEM is used in regard to a system that requires reliability, it is important to comprehensively detect input values that lead to malfunctions of a system (failures) during development and take appropriate measures. One conceivable solution is to install an input filter that controls the input to the DTEM, and to use separate software to process input values that may lead to failures. To develop the input filter, it is necessary to specify the filtering condition of the input value that leads to the malfunction of the system. Given that necessity, in this paper, we propose a method for formally verifying a DTEM and, according to the result of the verification, if an input value leading to a failure is found, extracting the range in which such an input value exists. The proposed method can comprehensively extract the range in which the input value leading to the failure exists; therefore, by creating an input filter based on that range, it is possible to prevent the failure occurring in the system. In this paper, the algorithm of the proposed method is described, and the results of a case study using a dataset of house prices are presented. On the basis of those results, the feasibility of the proposed method is demonstrated, and its scalability is evaluated.

Accomplishing the genotype-specific serodiagnosis of single and dual Trypanosoma cruzi infections by flow cytometry Chagas-Flow ATE-IgG2a.

The methods currently available for genotype-specific diagnosis of T. cruzi infection still present relevant limitations, especially to identify mixed infection. In the present investigation, we have evaluated the performance of Chagas-Flow ATE-IgG2a test for early and late differential diagnosis of single and dual genotype-specific T. cruzi infections. Serum samples from Swiss mice at early and late stages of T. cruzi infection were assayed in parallel batches for genotype-specific diagnosis of single (TcI, TcVI or TcII) and dual (TcI+TcVI, TcVI+TcII or TcII+TcI) infections. The intrinsic reactivity to TcI, TcVI and TcII target antigens, including amastigote (AI/AVI/AII), trypomastigote-(TI/TVI/TII) and epimastigote (EI/EVI/EII), at specific reverse of serum dilutions (500 to 64,000), was employed to provide reliable decision-trees for “early” vs “late”, “single vs “dual” and “genotype-specific” serology. The results demonstrated that selective set of attributes “EII 500/EI 2,000/AII 500” were able to provide high-quality accuracy (81%) to segregate early and late stages of T. cruzi infection. The sets “TI 2,000/AI 1,000/EII 1,000” and “TI 8,000/AII 32,000” presented expressive scores to discriminate single from dual T. cruzi infections at early (85%) and late stages (84%), respectively. Moreover, the attributes “TI 4,000/TVI 500/TII 1,000”, “TI 16,000/EI 2,000/EII 2,000/AI 500/TVI 500” showed good performance for genotype-specific diagnosis at early stage of single (72%) and dual (80%) T. cruzi infections, respectively. In addition, the attributes “TI 4,000/AII 1,000/EVI 1,000”, “TI 64,000/AVI 500/AI 2,000/AII 1,000/EII 4,000” showed moderate performance for genotype-specific diagnosis at late stage of single (69%) and dual (76%) T. cruzi infections, respectively. The sets of decision-trees were assembled to construct a sequential algorithm with expressive accuracy (81%) for serological diagnosis of T. cruzi infection. These findings engender new perspectives for the application of Chagas-Flow ATE-IgG2a method for genotype-specific diagnosis in humans, with relevant contributions for epidemiological surveys as well as clinical and post-therapeutic monitoring of Chagas disease.

Default Patterns in Seven EU Countries: A Random Forest Approach

This study uses the relatively new “random forest” (RF) approach, which is based on decision-tree analysis by combining the results of a large set of decision trees. RFs have so far been little used for default prediction but offer an interesting alternative to well-established default prediction techniques. Based on accounting data from 945,062 observed European firms from seven countries in 2010 and 1,019,312 firms in 2011, we provide evidence on the country-specific default patterns. Because of the strong imbalance of the data sets with regard to the solvency status, standard RF implementations have to be modified to allow the estimation of realistic default propensities. We find that by far most accurate out-of-sample default propensities can be obtained for Italy followed by Portugal and Spain and the least accurate for the UK and Finland. The debt ratio, rate of return on sales, dynamic gearing ratio, and the rate of return on assets are found to be the most important variables for default prediction. The variable importance rankings are rather country specific, pointing to heterogeneity in the default patterns across the countries studied.

Can Sentiment Indicators Signal Market Reversals?

In this study we use machine learning algorithm to test Amareos sentiment indicator's predictive power of market reversals. We then build and test a viable trading strategy.As input for the algorithm, we used eight market sentiment indicators (Anger, Anticipation, Disgust, Fear, Gloom, Joy, Optimism and Sentiment) on 20 major equity indices from January 1, 2005 to April 15, 2016.As the target output, we use a classification of the performance of the indices on the following 182 days - approximately six months - split between bottom, top and neutral days.Our learning algorithm is of the type called random forest. Through calibration on a training set composed of 64% of the data, we obtain a final set of decision trees, or forest. We then examine the out of sample accuracy of this forest on the remaining 36% of the data.As the accuracy on the test set is relatively high - a result that cannot be explained just by luck - we simulate a trading strategy based on the forest output. The resulting trading strategy produces strong performance, certainly much better than a simple buy and hold, even when adjusted for risk.

Fast HEVC Encoding Decisions Using Data Mining

The High Efficiency Video Coding standard provides improved compression ratio in comparison with its predecessors at the cost of large increases in the encoding computational complexity. An important share of this increase is due to the new flexible partitioning structures, namely the coding trees, the prediction units, and the residual quadtrees, with the best configurations decided through an exhaustive rate-distortion optimization (RDO) process. In this paper, we propose a set of procedures for deciding whether the partition structure optimization algorithm should be terminated early or run to the end of an exhaustive search for the best configuration. The proposed schemes are based on decision trees obtained through data mining techniques. By extracting intermediate data, such as encoding variables from a training set of video sequences, three sets of decision trees are built and implemented to avoid running the RDO algorithm to its full extent. When separately implemented, these schemes achieve average computational complexity reductions (CCRs) of up to 50% at a negligible cost of 0.56% in terms of Bjontegaard Delta (BD) rate increase. When the schemes are jointly implemented, an average CCR of up to 65% is achieved, with a small BD-rate increase of 1.36%. Extensive experiments and comparisons with similar works demonstrate that the proposed early termination schemes achieve the best rate-distortion-complexity tradeoffs among all the compared works.

Improving Classification Accuracy based on Random Forest Model with Uncorrelated High Performing Trees

Random forest can achieve high classification performance through a classification ensemble with a set of decision trees that grow using randomly selected subspaces of data. The performance of an ensemble learner is highly dependent on the accuracy of each component learner and the diversity among these components. In random forest, randomization would cause occurrence of bad trees and may include correlated trees. This leads to inappropriate and poor ensemble classification decision. In this paper an attempt has been made to improve the performance of the model by including only uncorrelated high performing trees in a random forest. Experimental results have shown that, the random forest can be further enhanced in terms of the classification accuracy. General Terms Random forest, Classification Accuracy, Uncorrelated trees.

Learning a Transferable World Model by Reinforcement Agent in Deterministic Observable Grid-World Environments

Reinforcement-based agents have difficulties in transferring their acquired knowledge into new different environments due to the common identities-based percept representation and the lack of appropriate generalization capabilities. In this paper, the problem of knowledge transferability is addressed by proposing an agent dotted with decision tree induction and constructive induction capabilities and relying on decomposable properties-based percept representation. The agent starts without any prior knowledge of its environment and of the effects of its actions. It learns a world model (the set of decision trees) that corresponds to the set of explicit action definitions predicting action effects in terms of agent’s percepts. Agent’s planning component uses predictions of the world model to chain actions via a breadth-first search. The proposed agent was compared to the Q-learning and Adaptive Dynamic Programming based agents and demonstrated better ability to achieve goals in static observable deterministic grid-world environments different from those in which it has learnt its world model.DOI: http://dx.doi.org/10.5755/j01.itc.41.4.915

Analysis of a random forests model

Random forests are a scheme proposed by Leo Breiman in the 2000's for building a predictor ensemble with a set of decision trees that grow in randomly selected subspaces of data. Despite growing in...

Active rule learning using decision tree for resource management in Grid computing

Grid computing is becoming a mainstream technology for large-scale resource sharing and distributed system integration. One underlying challenge in Grid computing is the resource management. In this paper, active rule learning is considered for resource management in Grid computing. Rule learning is very important for updating rules in an active database system. However, it is also very difficult because of a lack of methodology and support. A decision tree can be used in rule learning to cope with the problems arising in active semantic extraction, termination analysis of the rule set and rule updates. Also our aim in rule learning is to learn new attributes in rules, such as time and load balancing, in regard to instances of a real Grid environment that a decision tree can provide. In our work, a set of decision trees is built in parallel on training data sets based on the original rule set. Each learned decision tree can be reduced to a set of rules and thence conflicting rules can be resolved. Results from cross validation experiments on a data set suggest this approach may be effectively applied for rule learning.

Input Decimated Ensemble based on Neighborhood Preserving Embedding for spectrogram classification

In this work a novel technique for building ensembles of classifiers for spectrogram classification is presented. We propose a simple approach for classifying signals from a large database of plant echoes, these echoes are highly complex stochastic signals, anyway their spectrograms contain enough information for extracting a good set of features for training the proposed ensemble of classifiers. The proposed ensemble of classifiers is a novel modified version of a recent feature transform based ensemble method: the Input Decimated Ensemble. In the proposed variant different subsets of randomly extracted training patterns are used to create a set of different Neighborhood Preserving Embedding subspace projections. These feature transformations are applied to the whole dataset and a set of decision trees are trained using these transformed spaces. Finally, the scores of this set of classifiers are combined by sum rule. Experiments carried out on a yet proposed dataset show the superiority of this method with respect to other approaches. The proposed approach outperforms the yet proposed, for the tested dataset, combination of principal component analysis and support vector machine (SVM). Moreover, we show that the fusion between the proposed ensemble and the system based on SVM outperforms both the stand-alone methods.

An Efficient Algorithm for Generating Generalized Decision Forests

A shortcoming of univariate decision tree learners is that they do not learn intermediate concepts and select only one of the input features in the branching decision at each intermediate tree node. It has been empirically demonstrated that cascading other classification methods, which learn intermediate concepts, with decision tree learners can alleviate such representational bias of decision trees and potentially improve classification performance. However, a more complex model that fits training data better may not necessarily perform better on unseen data, commonly referred to as the overfitting problem. To find the most appropriate degree of such cascade generalization, a decision forest (i.e., a set of decision trees with other classification models cascaded to different degrees) needs to be generated, from which the best decision tree can then be identified. In this paper, the authors propose an efficient algorithm for generating such decision forests. The algorithm uses an extended decision tree data structure and constructs any node that is common to multiple decision trees only once. The authors have empirically evaluated the algorithm using 32 data sets for classification problems from the University of California, Irvine (UCI) machine learning repository and report on results demonstrating the efficiency of the algorithm in this paper.

A decision tree system for finding genes in DNA.

MORGAN is an integrated system for finding genes in vertebrate DNA sequences. MORGAN uses a variety of techniques to accomplish this task, the most distinctive of which is a decision tree classifier. The decision tree system is combined with new methods for identifying start codons, donor sites, and acceptor sites, and these are brought together in a frame-sensitive dynamic programming algorithm that finds the optimal segmentation of a DNA sequence into coding and noncoding regions (exons and introns). The optimal segmentation is dependent on a separate scoring function that takes a subsequence and assigns to it a score reflecting the probability that the sequence is an exon. The scoring functions in MORGAN are sets of decision trees that are combined to give a probability estimate. Experimental results on a database of 570 vertebrate DNA sequences show that MORGAN has excellent performance by many different measures. On a separate test set, it achieves an overall accuracy of 95 %, with a correlation coefficient of 0.78, and a sensitivity and specificity for coding bases of 83 % and 79%. In addition, MORGAN identifies 58% of coding exons exactly; i.e., both the beginning and end of the coding regions are predicted correctly. This paper describes the MORGAN system, including its decision tree routines and the algorithms for site recognition, and its performance on a benchmark database of vertebrate DNA.