Decision Tree Logic Research Articles

Gleaning data from disaster: A hospital-based data mining method to study all-hazard triage after a chemical disaster

To describe the methods of evaluating currently available triage models for their efficacy in appropriately triaging the surge of patients after an all-hazards disaster. A method was developed for evaluating currently available triage models using extracted data from medical records of the victims from the Graniteville chlorine disaster. On January 6, 2005, a freight train carrying three tanker cars of liquid chlorine was inadvertently switched onto an industrial spur in central Graniteville, SC. The train then crashed into a parked locomotive and derailed. This caused one of the chlorine tankers to rupture and immediately release ~60 tons of chlorine. Chlorine gas infiltrated the town with a population of 7,000. This research focuses on the victims who received emergency care in South Carolina. With our data mapping and decision tree logic, the authors were successful in using the available extracted clinical data to estimate triage categories for use in our study. The methodology outlined in this article shows the potential use of well-designed secondary analysis methods to improve mass casualty research. The steps are reliable and repeatable and can easily be extended or applied to other disaster datasets.

Integrating individual trip planning in energy efficiency – Building decision tree models for Danish fisheries

Danish fishermen have provided information on dynamics in their fuel consumption, running costs, and fishing patterns through a web-based questionnaire. This detailed documentation of the fishing practices is used in spatial modelling tools to improve advice and research for fisheries. The tools integrate detailed information on vessel distribution, catch and fuel consumption for different fisheries with a detailed resource distribution of targeted stocks from research surveys to evaluate the optimum consumption and efficiency to reduce fuel costs and the costs of displacement of effort. The energy efficiency for the value of catch per unit of fuel consumed is analysed by merging the questionnaire, logbook and VMS (vessel monitoring system) information. Logic decision trees and conditional behaviour probabilities are established from the responses of fishermen regarding a range of sequential hypothetical conditions influencing their trip decisions, covering the duration of fishing time, choice of fishing ground(s), when to stop fishing and return to port, and the choice of the port for landing. Fleet-based energy and economy efficiency are linked to the decision (choice) dynamics. Larger fuel-intensive but efficient vessels conducting pelagic or industrial fishing are more inclined to base their decision on fish price only, while numerous smaller and less efficient vessels conducting demersal mixed or crustacean fishery usually consider other flexible factors, e.g., the potential for a large catch, weather, previous knowledge and experience, and the distance to/from port, which affect the number and duration of trips and the fuel consumption. Integration of the results into our recently developed spatially explicit individual-based fishing vessel model (IBM) incorporate the variability and predict the adaptations of individual fishermen to resource availability dynamics, increasing fuel prices, changes in regulations, and the consequences of socioeconomic external pressures on harvested stocks. A new methodology is described here to obtain quantitative information on the fishermen's micro-scale decisions initially required.

Instant spectral assignment for advanced decision tree-driven mass spectrometry

We have developed and implemented a sequence identification algorithm (inSeq) that processes tandem mass spectra in real-time using the mass spectrometer's (MS) onboard processors. The inSeq algorithm relies on accurate mass tandem MS data for swift spectral matching with high accuracy. The instant spectral processing technology takes ∼16ms to execute and provides information to enable autonomous, real-time decision making by the MS system. Using inSeq and its advanced decision tree logic, we demonstrate (i) real-time prediction of peptide elution windows en masse (∼3min width, 3,000 targets), (ii) significant improvement of quantitative precision and accuracy (~3x boost in detected protein differences), and (iii) boosted rates of posttranslation modification site localization (90% agreement in real-time vs. offline localization rate and an approximate 25% gain in localized sites). The decision tree logic enabled by inSeq promises to circumvent problems with the conventional data-dependent acquisition paradigm and provides a direct route to streamlined and expedient targeted protein analysis.

A Robust Error-Based Rain Estimation Method for Polarimetric Radar. Part II: Case Study

AbstractRainfall estimation using polarimetric radar involves the combination of a number of estimators with differing error characteristics to optimize rainfall estimates at all rain rates. In Part I of this paper, a new technique for such combinations was proposed that weights algorithms by the inverse of their theoretical errors. In this paper, the derived algorithms are validated using the “CP2” polarimetric radar in Queensland, Australia, and a collocated rain gauge network for two heavy-rain events during November 2008 and a larger statistical analysis that is based on data from between 2007 and 2009. Use of a weighted combination of polarimetric algorithms offers some improvement over composite methods that are based on decision-tree logic, particularly at moderate to high rain rates and during severe-thunderstorm events.

A Robust Error-Based Rain Estimation Method for Polarimetric Radar. Part I: Development of a Method

AbstractThe algorithms used to estimate rainfall from polarimetric radar variables show significant variance in error characteristics over the range of naturally occurring rain rates. As a consequence, to improve rainfall estimation accuracy using polarimetric radar, it is necessary to optimally combine a number of different algorithms. In this study, a new composite method is proposed that weights the algorithms by the inverse of their theoretical error. A number of approaches are discussed and are investigated using simulated radar data calculated from disdrometer measurements. The resultant algorithms show modest improvement over composite methods based on decision-tree logic—in particular, at rain rates above 20 mm h−1.

How GAAP and accounting treatments influence property management

PurposeThe purpose of this study is to identify how different Generally Accepted Accounting Principles (GAAP) influence property management. The study is based on two basic accounting principles for the valuation of assets: fair value and historical cost. The study focuses on land and buildings as a main part of the total fixed assets of a company. It uses the framework of the Greek real estate market as an experimental setting where the principles of historic cost and fair value accounting can be compared.Design/methodology/approachThe topic is approached using an integration of fixed assets into four main portfolio categories: own used; investments; held for sale assets; and inventories. According to this framework the study examines the accounting treatments under International Financial Reporting Standards (IFRS), US GAAP and Greek GAAP for each portfolio transaction and analyses the impact of accounting entries to equity and profit and loss account.FindingsThe study results to a comparative analysis of the different studied GAAP and tries to establish a purchase price allocation method for property acquisition.Originality/valueThe contribution of this article is that it surveys principles, literature and practice about the above issues from a critical perspective, and presents a way to managing and monitoring real estate investments, using logical decision trees, from an accounting point of view.

Detection and Extraction of Videos using Decision Trees

This paper addresses a new multimedia data mining framework for the extraction of events in videos by using decision tree logic. The aim of our DEVDT (Detection and Extraction of Videos using Decision Trees) system is for improving the indexing and retrieval of multimedia information. The extracted events can be used to index the videos. In this system we have considered C4.5 Decision tree algorithm [3] which is used for managing both continuous and discrete attributes. In this process, firstly we have adopted an advanced video event detection method to produce event boundaries and some important visual features. This rich multi-modal feature set is filtered by a pre-processing step to clean the noise as well as to reduce the irrelevant data. This will improve the performance of both Precision and Recall. After producing the cleaned data, it will be mined and classified by using a decision tree model. The learning and classification steps of this Decision tree are simple and fast. The Decision Tree has good accuracy. Subsequently, by using our system we will reach maximum Precision and Recall i.e. we will extract pure video events effectively and proficiently.

A new method of estimating the ratio between in situ rock stresses and tectonics based on empirical and probabilistic analyses

This paper describes a new procedure for assessing the ratio between in situ stresses in rock masses by means of K ( K = σ H / σ v, being σ H and σ v principal stress) and tectonics for purposes of engineering geology and rock mechanics. The method combines the use of the logic decision tree and the empirical relationship between the Tectonic Stress Index, TSI, and a series of K in situ values obtained from an extensive database. The decision tree considers geological and geophysical factors affecting stress magnitudes both on the regional and local scale. The TSI index is defined by geological and geomechanical parameters. The method proposed provides an assessment of the magnitude of horizontal stresses of tectonic origin. Results for several regions of Europe are presented and the possible applications of the procedure are discussed.

A Fuzzy Decision Tree for Fault Classification

In plant accident management, the control room operators are required to identify the causes of the accident, based on the different patterns of evolution of the monitored process variables thereby developing. This task is often quite challenging, given the large number of process parameters monitored and the intense emotional states under which it is performed. To aid the operators, various techniques of fault classification have been engineered. An important requirement for their practical application is the physical interpretability of the relationships among the process variables underpinning the fault classification. In this view, the present work propounds a fuzzy approach to fault classification, which relies on fuzzy if-then rules inferred from the clustering of available preclassified signal data, which are then organized in a logical and transparent decision tree structure. The advantages offered by the proposed approach are precisely that a transparent fault classification model is mined out of the signal data and that the underlying physical relationships among the process variables are easily interpretable as linguistic if-then rules that can be explicitly visualized in the decision tree structure. The approach is applied to a case study regarding the classification of simulated faults in the feedwater system of a boiling water reactor.

Implementation and Comparison of Reservoir Operation Strategies for the Hoa Binh Reservoir, Vietnam using the Mike 11 Model

Hoa Binh is the largest reservoir in Vietnam. It has been operated since 1990 with the main purposes of flood control in the Red River basin and hydropower generation. Because these different purposes always cause conflicts and disputes during the flood season, it is desirable to improve the current operational regulations of the reservoir. In this paper, the operation rules of the reservoir are analysed by applying the Mike 11 river modelling tool. The model set up includes the main rivers and tributaries of the Red River basin and a logical decision tree defining the reservoir regulation. These strategies define the reservoir release as function of the time of the year, the actual reservoir stage, and the water level forecast at Hanoi. A data set consisting of twenty years of flood season data was used to evaluate the control strategies with respect to flood control and hydropower generation. The reservoir operation using the complete control system and the current as well as alternative regulation strategies has been evaluated and compared to the actual operation practice. Results showed that the implemented control system performs better than the actual operation. In addition, lowering the target downstream water level for flood control improves the operation with respect to both flood protection and hydropower generation. An alternative strategy where the target water level in the reservoir is increased can improve hydropower generation but at the expense of a reduced flood protection.

A multimodal data mining framework for soccer goal detection based on decision tree logic

In this paper, we propose a new multimedia data mining framework for the extraction of soccer goal events in soccer videos by utilising both multimodal analysis and decision tree logic. The extracted events can be used to index the soccer videos. We first adopt an advanced video shot detection method to produce shot boundaries and some important visual features. Then, the visual/audio features are extracted for each shot at different granularities. This rich multimodal feature set is then filtered by a pre-filtering step in order to clean the noise as well as to reduce the irrelevant data. A decision tree model is built upon the cleaned data set and is used to classify the goal shots. We also present the experimental results for the proposed framework, which indicate the performance of the framework for soccer goal extraction.

Psyconsultant I: A DSM-IV-Based Screening Tool For Adult Psychiatric Disorders In Indian Rural Health Center

Computerized diagnostic tools have made havoc in medical science for last few decades. With the help of computerized expert tool, psychiatric illnesses can be diagnosed fast, with good accuracy rate and free from human bias. Moreover, in Indian rural health sectors, where psychiatrists are less in number, computerized screening tool could be extremely helpful for the general physicians to evaluate mental patients. The present tool, PsyConsaltant I has a long questionnaire, covering eleven broad domains of psychiatric illnesses and their sub domains. The questionnaire is set with the guidance of qualified psychiatrists to incorporate clinical experiences along with DSM-IV nosology and is put together in a computer program using logical decision tree. Finally the performance of the tol has been tested on a large set of patients and its diagnostic accuracy rate is found to be over 55% (with some variations in individual domains) and having statistical significance level 0.05.

Reaction intermediates in organic chemistry — The “big picture”

An overview of the discovery of reaction intermediates and associated concepts in physical organic chemistry is presented. Particular attention is paid to chronology of ideas, frequency of occurrence of reaction intermediates in the library of organic reactions used in organic synthesis, and the lexicon of scientific terms used in the language of physical organic chemistry. General logic decision trees are presented for the unique or near unique identification of reaction intermediates based on experimental techniques and common patterns of reactivity documented in the literature over the last century. Contributions made by scientists working in laboratories at Canadian universities and at the National Research Council of Canada are noted throughout.Key words: physical organic chemistry, mechanistic chemistry, reaction intermediates.

Current Icing Potential: Algorithm Description and Comparison with Aircraft Observations

AbstractThe “current icing potential” (CIP) algorithm combines satellite, radar, surface, lightning, and pilot-report observations with model output to create a detailed three-dimensional hourly diagnosis of the potential for the existence of icing and supercooled large droplets. It uses a physically based situational approach that is derived from basic and applied cloud physics, combined with forecaster and onboard flight experience from field programs. Both fuzzy logic and decision-tree logic are applied in this context. CIP determines the locations of clouds and precipitation and then estimates the potential for the presence of supercooled liquid water and supercooled large droplets within a given airspace. First developed in the winter of 1997/98, CIP became an operational National Weather Service and Federal Aviation Administration product in 2002, providing real-time diagnoses that allow users to make route-specific decisions to avoid potentially hazardous icing. The CIP algorithm, its individual components, and the logic behind them are described.

Guidelines for the Systematic Treatment of the Depressed Patient

LARRY E. BEUTLER, JOHN F. CLARKIN, AND BRUCE BONGAR: Guidelines for Systematic Treatment of Depressed Patient. Oxford University Press, New York, 2000, 455 pp., ISBN 0-19-510530-3. Guidelines for Systematic Treatment of Depressed Patient is first in a series of guidebooks whose goal is to provide mental health practitioners a framework to treat individuals with depressive disorders. It is written by three highly respected academic psychologists: Larry E. Beutler, editor of Journal of Clinical Psychology, and Professor at UC Santa Barbara; John F Clarkin, Professor at Cornell's Weill Medical College; and Bruce Bongar, Professor at Pacific Graduate School of Psychology. The volume is organized into six parts. In its 14 chapters work undertakes a scholarly review, largely of psychology literature, of state of art of treatment of depression. It presents senior author's life work, developing his theoretical approach to evaluation, treatment, and ongoing assessment of individuals with depression who seek comprehensive psychotherapeutic treatment from mental health professionals. Part I sets context by reviewing need for treatment guidelines in mental health and two already established U.S. depression guidelines: The Agency for Health Care Policy and Research (AHCPR) 1993 Depression Guidelines for Primary Care and American Psychiatric Association's 1993 Practice Guidelines for major depressive disorder in adults. Part II (one third of text) develops authors' theoretical framework that guidelines for treatment-relevant assessment must be viewed through lens of a traitlike dimensional (as opposed to categorical) approach to patients. In Chapter three, one of book's two strongest chapters, using a comprehensive and scholarly review of literature (including 57 pages of Tabular review of hundreds of studies) authors identify six patient qualities/dimensions critical for selecting and assessing treatment: functional impairment, subjective distress, perceived social support, problem complexity/chronicity, level of resistance, and coping style. Throughout rest of work, ten general classes of psychotherapeutic interventions are discussed in context of these six dimensions. Chapter four introduces authors' theoretical framework, Systematic Treatment Selection (STS), on which proposed comprehensive interventions are based. The STS model and two illustrative case examples are presented along with: (1) a comprehensive, but impractical, method of identifying and assessing dimensions, relying heavily on psychological testing; and (2) a system for using decision-tree logic to derive treatment planning, with available computer-assisted procedures. Despite authors' disclaimer that, the current guidelines were not developed to comply with political agenda or dictating philosophies of any specific group (p. …

Operation experience of a solar- and wind-powered desalination demonstration plant

The present work outlines the designing, erection and operation process of a stand-alone desalination plant powered by both solar photovoltaic and wind energy. Such a plant will serve small isolated communities in remote areas devoid of water resources. A specially customized code was built to simulate the operation of the installation in order to allow appropriate choice of components specifications. Site meteorological data were used to enhance prediction capabilities. The code continuously updates the instantaneous water lever of the reservoir as well as the current state of charge of the accumulators. Depending on these two variables, a logical decision tree is built to decide whether the cumulated wind and solar energy production can satisfy the load of the plant or additional energy must be provided from the accumulators or an auxiliary diesel engine generator. The process control system for such an installation must allow for operation in isolated areas where qualified maintenance personnel are scarce or remote. These are special considerations regarding the design philosophy in order to reach a state as close as possible to a maintenance-free system. In view of this consideration, several layers of back-up were built into the system such as a diesel generator (whose use is to be kept to a minimum). Also, the system has been designed to operate at about a 33% service factor. Two-day battery storage autonomy has also been provided. The desalination plant uses reverse osmosis technology. The plant has a maximum product capacity of 9 m 3/d in view of future needs, even though it is designed to currently produce only 3 m 3/d. The inlet water is to be provided from on-site brackish water wells. The local water quality is approximately 3500–5000 ppm corresponding to brackish water. The system has been designed based on the premise that the average on-site wind velocity is about 4–5 m/s and an isolation level of about 5–5.5 kWh/m 2/d. The expected life-span of the plant is about 15 years. The system was successfully erected and has been continuously operated producing 3 m 3/d. Experimental measurements are now in progress, and a comparison to theoretical predictions is presented. The time schedule for the whole project consisted of 6–8 months, including many changes required during construction.

Modified region decomposition method and optimal depth decision tree in the recognition of non-uniform sized characters – An experimentation with Kannada characters

In contrast to English alphabets, some characters in Indian languages such as Kannada, Hindi, Telugu may have either horizontal or vertical or both the extensions making it difficult to enclose every such character in a standard rectangular grid as done quite often in character recognition research. In this work, an improved method is proposed for the recognition of such characters (especially Kannada characters), which can have spread in vertical and horizontal directions. The method uses a standard sized rectangle which can circumscribe standard sized characters. This rectangle can be interpreted as a two-dimensional, 3×3 structure of nine parts which we define as bricks. This structure is also interpreted as consecutively placed three row structures of three bricks each or adjacently placed three column structures of three bricks each.It is obvious that non-uniform sized characters cannot be contained within the standard rectangle of nine bricks. The work presented here proposes to take up such cases. If the character has horizontal extension, then the rectangle is extended horizontally by adding one column structure of three bricks at a time, until the character is encapsulated. Likewise, for vertically extended characters, one row structure is added at a time. For the characters which are smaller than the standard rectangle, one column structure is removed at a time till the character fits in the shrunk rectangle. Thus, the character is enclosed in a rectangular structure of m×n bricks where m⩾3 and n⩾1. The recognition is carried out intelligently by examining certain selected bricks only instead of all mn bricks. The recognition is done based on an optimal depth logical decision tree developed during the Learning phase and does not require any mathematical computation.

Top-down induction of first order logical decision trees

Although topdown induction of decision trees is a very popular induction method, up till now it has mainly been used for propositional learning relational decision tree learners are scarce. This di...

Scaling Up Inductive Logic Programming by Learning from Interpretations

When comparing inductive logic programming (ILP) and attribute-value learning techniques, there is a trade-off between expressive power and efficiency. Inductive logic programming techniques are typically more expressive but also less efficient. Therefore, the data sets handled by current inductive logic programming systems are small according to general standards within the data mining community. The main source of inefficiency lies in the assumption that several examples may be related to each other, so they cannot be handled independently. Within the learning from interpretations framework for inductive logic programming this assumption is unnecessary, which allows to scale up existing ILP algorithms. In this paper we explain this learning setting in the context of relational databases. We relate the setting to propositional data mining and to the classical ILP setting, and show that learning from interpretations corresponds to learning from multiple relations and thus extends the expressiveness of propositional learning, while maintaining its efficiency to a large extent (which is not the case in the classical ILP setting). As a case study, we present two alternative implementations of the ILP system TILDE (Top-down Induction of Logical DEcision trees): TILDEclassic, which loads all data in main memory, and TILDELDS, which loads the examples one by one. We experimentally compare the implementations, showing TILDELDS can handle large data sets (in the order of 100,000 examples or 100 MB) and indeed scales up linearly in the number of examples.

Top-down induction of first-order logical decision trees

A first-order framework for top-down induction of logical decision trees is introduced. The expressivity of these trees is shown to be larger than that of the flat logic programs which are typically induced by classical ILP systems, and equal to that of first-order decision lists. These results are related to predicate invention and mixed variable quantification. Finally, an implementation of this framework, the TILDE system, is presented and empirically evaluated.