This paper first provides an overview of data preprocessing, focusing on problems of real world data. These are primarily problems that have to be carefully understood and solved before any data analysis process can start. The paper discusses in detail two main reasons for performing data preprocessing: i problems with the data and ii preparation for data analysis. The paper continues with details of data preprocessing techniques achieving each of the above mentioned objectives. A total of 14 techniques are discussed. Two examples of data preprocessing applications from two of the most data rich domains are given at the end. The applications are related to semiconductor manufacturing and aerospace domains where large amounts of data are available, and they are fairly reliable. Future directions and some challenges are discussed at the end.

/pdf/data-preprocessing-and-intelligent-data-analysis-482v7nfhdy.pdf

Data Preprocessing and Intelligent Data Analysis

Compared with forces occurring in soil mechanics problems in civil engineering, the forces that are applied to soil in farming operations generally have a short duration, less than a few seconds, a small loaded area, no more than a few square decimeters, and small intensities, 10 bar being a high value On the other hand, soil properties vary widely between those of a weak mud and a stone-like dry soil Tillage and related applications of force to soil are practiced worldwide in farming Tillage operations are performed on one hectare of land for every three human beings This means that for the food production for each individual daily, something like one cubic meter of soil is stirred, or about 20 times his body weight Theoretical knowledge of this most common human activity, which largely determines the surface shape of the fertile part of the earth, is still very limited In this book the authors have tried to give an outline of the present state of the art One of the starting points was a course in soil dynamics taught by the authors at the Agricultural University at Wageningen, The Netherlands We hope to reach interested readers who have no more theoretical knowledge than high school level, as well as readers who want to go beyond the level of a third year university student For the chapter on wheels and tires we received substantial support from F G J Tijink of the Tillage Laboratory at Wageningen

Agricultural Soil Mechanics

In this paper, five major shifts in engineering education are identified. During the engineering science revo- lution, curricula moved from hands-on practice to mathemat- ical modeling and scientific analyses. The first shift was initiated by engineering faculty members from Europe; accel- erated during World War II, when physicists contributed mul- tiple engineering breakthroughs; codified in the Grinter report; and kick-started by Sputnik. Did accreditation hinder curricular innovations? Were engineering graduates ready for practice? Spurred by these questions, the Accreditation Board for Engineering and Technology (ABET) required engineering programs to formulate outcomes, systematically assess achievement, and continuously improve student learning. The last three shifts are in progress. Since the engineering science revolution may have marginalized design, a distinctive feature of engineering, faculty members refocused attention on cap- stone and first-year engineering design courses. However, this third shift has not affected the two years in between. Fourth, research on learning and education continues to influence engineering education. Examples include learning outcomes and teaching approaches, such as cooperative learning and inquiry that increase student engagement. In shift five, tech- nologies (e.g., the Internet, intelligent tutors, personal compu- ters, and simulations) have been predicted to transform education for over 50 years; however, broad transformation has not yet been observed. Together, these five shifts characterize changes in engineering education over the past 100 years.

Five Major Shifts in 100 Years of Engineering Education The authors discuss what has reshaped, or is currently reshaping, engineering education over the past 100 years up until the current emphasis on design, learning, and social-behavioral sciences research and the role of technology.

In this paper, five major shifts in engineering education are identified. During the engineering science revolution, curricula moved from hands-on practice to mathematical modeling and scientific analyses. The first shift was initiated by engineering faculty members from Europe; accelerated during World War II, when physicists contributed multiple engineering breakthroughs; codified in the Grinter report; and kick-started by Sputnik. Did accreditation hinder curricular innovations? Were engineering graduates ready for practice? Spurred by these questions, the Accreditation Board for Engineering and Technology (ABET) required engineering programs to formulate outcomes, systematically assess achievement, and continuously improve student learning. The last three shifts are in progress. Since the engineering science revolution may have marginalized design, a distinctive feature of engineering, faculty members refocused attention on capstone and first-year engineering design courses. However, this third shift has not affected the two years in between. Fourth, research on learning and education continues to influence engineering education. Examples include learning outcomes and teaching approaches, such as cooperative learning and inquiry that increase student engagement. In shift five, technologies (e.g., the Internet, intelligent tutors, personal computers, and simulations) have been predicted to transform education for over 50 years; however, broad transformation has not yet been observed. Together, these five shifts characterize changes in engineering education over the past 100 years.

Five Major Shifts in 100 Years of Engineering Education

The proportionate mortality experience of 173 decedents exposed primarily to lead arsenate and calcium arsenate was compared with that of 1,809 decedents not exposed to those compounds. A significant increase in respiratory cancer was found among the exposed employees. The relationship between cumulative arsenic exposure and the ratio of observed to expected respiratory malignancy deaths was estimated by means of a least squares approach. The predicted ratio was 7:1 for individuals exposed for a period of more than eight years to compounds that contained an equivalent level of 1 mg/cu m arsenic. In the more heavily exposed individuals, an excess of respiratory cancer was observed 35+ years after the initial exposure. Observations based on the proportionate study were supported by an analysis of the same target population, employing the prospective study in retrospect approach.

Respiratory Cancer and Occupational Exposure to Arsenicals

Similitude in engineering

Hardness tests are widely used for determining comparative hardness numbers for metals, because of their simplicity and rapidity of operation. If appropriate charts or equations are known a given hardness number can be converted to other hardness numbers or to some of the mechanical properties.

https://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1057&context=ameslab_iscreports

Correlation of Vickers hardness number, modulus of elasticity, and the yield strength for ductile metals

Instrumental neutron activation, in which no chemical separation is involved, has been applied to the determination of arsenic content in American cigarettes. The technique has the merits of being nondestructive and having no significant interferences. Fifty‐two samples of 33 brands of cigarettes have been analyzed. The results show that one gram of tobacco contains, on the average, 7.69 ± 0.48 μg of arsenic, ranging from 5.5 ± 0.34 μg to 12.8 ± 0.79 μg.

Determination of arsenic content of american cigarettes by neutron activation analysis

Similitude requirements for predicting the response of dynamically loaded buried structures by means of scale models were developed by dimensional analysis and were checked experimentally. Hollow circular and square aluminum tubes were buried in dry compacted Ottawa sand and loaded dynamically by dropping a cylindrical weight on the soil surface. The strains induced on the inner surfaces of the tubes were measured using SR-4 strain gages. Complete strain-time curves and data indicating the variation of peak strain and rise time with depth of burial are presented. Reasonable correlation is obtained between both the peak strain data and the complete strain-time curves for the several structures tested. In general, the study demonstrates the feasibility and value of the use of scale models for the prediction of the dynamic response of underground structures.

Dynamic Similitude of Underground Structures

A method was developed and investigated for directly measuring the shutdown margin in a reactor operating at a subcritical steady state. Measurements were made on the Iowa State University UTR-10 reactor with conditions ranging from almost critical to fully shutdown. The method is based on the fact that the frequency dependent transfer function of the reactor is a measure of the response of the reactor to an input disturbance. The magnitude of the transfer function of a subcritical reactor depends on the frequency of the disturbance and the steady-state multiplication factor. Since the multiplication factor is a measure of the reactivity, the amplitude of the transfer function is a direct indication of the shutdown margin. A sinusoidal disturbance was introduced by means of an oscillator operating in the center of the core. The oscillator was operated at a fixed frequency of 10 cps in order to eliminate the frequency dependence. The fluctuation was measured by a BF3 detector located against the outside of the core opposite the oscillator. The signal emitted by the detector contained bombardment noise and instrument noise in addition to the desired information. The signal was passed through a band-pass filter and was then cross-correlated with a pure sine wave to obtain the transfer function. A plot of the transfer function vs. the multiplication factor was a smooth function from which the multiplication factor could be found with a confidence of ± 0.04% near critical and with a confidence of ± 0.17% in the fully shutdown condition.

https://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=4956&context=rtd

Glenn Murphy

Papers

Similitude in engineering

Correlation of Vickers hardness number, modulus of elasticity, and the yield strength for ductile metals

Determination of arsenic content of american cigarettes by neutron activation analysis

Dynamic Similitude of Underground Structures

Measurement of Shutdown Margin