다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Data Mining - Concepts and Techniques.

Data Mining Practical Machine Learning Tools and Techniques

Hundreds of different types of coatings are used to protect a variety of structural engineering materials from corrosion, wear, and erosion, and to provide lubrication and thermal insulation. Of all these, thermal barrier coatings (TBCs) have the most complex structure and must operate in the most demanding high-temperature environment of aircraft and industrial gas-turbine engines. TBCs, which comprise metal and ceramic multilayers, insulate turbine and combustor engine components from the hot gas stream, and improve the durability and energy efficiency of these engines. Improvements in TBCs will require a better understanding of the complex changes in their structure and properties that occur under operating conditions that lead to their failure. The structure, properties, and failure mechanisms of TBCs are herein reviewed, together with a discussion of current limitations and future opportunities.

http://science.sciencemag.org/content/sci/296/5566/280.full.pdf

Thermal Barrier Coatings for Gas-Turbine Engine Applications

A model and the necessary database for predicting the response of powder metallurgy steels to heat treatment are presented and discussed. The model is based on a modification of the commercially available software DANTE coupled to the finite element analysis software ABAQUS. The model requires an extensive database that includes temperature- and porosity-dependent phase transformation kinetics, and temperature- and porosity-dependent phase-specific mechanical, physical, and thermal properties of the steel. This data is developed for FL-4065 PM alloy and is used in the model to predict dimensional change, distortion, residual stresses, and type and quantity of metallurgical phases present in the microstructure of a typical PM component after the component is subjected to a specified heat treatment schedule. The model predictions are found to be in very good agreement with measured values.

FEA model for predicting the response of powder metallurgy steel components to heat treatment

A new mathematical model for analyzing hydrogen permeation in solids, in which surface effects and traps influence hydrogen transport, is presented and solved. The new model combines the McNabb and Foster equations for diffusion with concomitant trapping[1] and a surface-limited mass-transfer boundary condition. An important result of the new model is the introduction of a new variable,hm, which is defined as the surface-limited mass-transfer coefficient. Thehm coefficient can account for all possible surface effects and may be experimentally evaluated.

Modeling surface effects on hydrogen permeation in metals

Quenching heat treatment in a liquid medium is a very complex heat transfer process. Heat extraction from the part surface occurs through several different heat transfer mechanisms in distinct temperature ranges, namely, film boiling, partial film boiling (i.e. transition), nucleate boiling and convection. The maximum heat transfer occurs during the nucleate boiling stage. Experimental study shows that, the effective surface heat transfer coefficient varies more than two orders of magnitude with the temperature during the quenching. For quenching process simulation, accurate prediction of the time-temperature history and microstructure evolution within the part largely depends on the accuracy of the boundary condition supplied. The heat transfer coefficient is the most important boundary condition for process simulation. This study focuses on creating a database of heat transfer coefficients for various liquid quenchant-metallic alloy combinations through experimentation using three different quench probes. This database is a web-based tool for use in quench process simulation. It provides at-a-glance information for quick and easy analysis and sets the stage for a Decision Support System (DSS) and Data Mining for heat-treating process.

Heat transfer coefficients for quenching process simulation

The crystallographic texture of EB-PVD TBCs has been experimentally determined by pole figure analysis and X-ray diffraction. It was found that the TBC coating deposited on a stationary flat surface directly above an ingot or between the two ingots and off center exhibited single crystal texture. Coatings deposited at the far corners of the coating chamber revealed a fiber texture with 〈111〉 column growth direction. The Bragg–Brentano X-ray diffraction analysis revealed a variety of textures (i.e. crystallographic planes that are parallel to the substrate surface) with the most predominant planes being the (220), (311), (420) and (620).

Crystallographic texture of EB-PVD TBCs deposited on stationary flat surfaces in a multiple ingot coating chamber as a function of chamber position

In mining graphical data the default Euclidean distance is often used as a notion of similarity. However this does not adequately capture semantics in our targeted domains, having graphical representations depicting results of scientific experiments. It is seldom known a-priori what other distance metric best preserves semantics. This motivates the need to learn such a metric. A technique called LearnMet is proposed here to learn a domain-specific distance metric for graphical representations. Input to LearnMet is a training set of correct clusters of such graphs. LearnMet iteratively compares these correct clusters with those obtained from an arbitrary but fixed clustering algorithm. In the first iteration a guessed metric is used for clustering. This metric is then refined using the error between the obtained and correct clusters until the error is below a given threshold. LearnMet is evaluated rigorously in the Heat Treating domain which motivated this research. Clusters obtained using the learned metric and clusters obtained using Euclidean distance are both compared against the correct clusters over a separate test set. Our results show that the learned metric provides better clusters.

/pdf/learning-semantics-preserving-distance-metrics-for-3a3qfnqgav.pdf

Richard D. Sisson

Papers

FEA model for predicting the response of powder metallurgy steel components to heat treatment

Modeling surface effects on hydrogen permeation in metals

Heat transfer coefficients for quenching process simulation

Crystallographic texture of EB-PVD TBCs deposited on stationary flat surfaces in a multiple ingot coating chamber as a function of chamber position

Learning semantics-preserving distance metrics for clustering graphical data