By Shahab D. Mohaghegh (Ed.), Saud M. Al-Fattah (Ed.), Andrei S. Popa (Ed.)
Read or Download Artificial Intelligence & Data Mining Applications in the E&P Industry PDF
Similar mining books
Die Herausgeber sind Professoren der Wirtschaftsinformatik an den Universitäten Mannheim und Marburg; neben den Herausgebern haben als weitere Autoren ehemalige Hochschul-Absolventen mitgewirkt, die nun als Entscheider und Praktiker in namhaften Firmen tätig sind.
This publication covers the elemental strategies of knowledge mining, to illustrate the opportunity of collecting huge units of knowledge, and studying those info units to realize beneficial company realizing. The booklet is geared up in 3 elements. half I introduces ideas. half II describes and demonstrates easy facts mining algorithms.
Photocopy caliber - yet readable
- Discrimination and Privacy in the Information Society: Data Mining and Profiling in Large Databases
- Statistics and Data Analysis in Geochemical Prospecting
- Formation Testing: Pressure Analysis
- Guidelines for Open Pit Slope Design
- Seismic Amplitude Inversion in Reflection Tomography (Handbook of Geophysical Exploration: Seismic Exploration) (Vol 33)
- Minerals Handbook 1996–97: Statistics & Analyses of the World’s Minerals Industry
Additional resources for Artificial Intelligence & Data Mining Applications in the E&P Industry
A-I4) Eq. A-I4 gives the weight change in a link between Layer j and an output layer. 1/ usually is called the learning rate and takes any value between 0 and 1. The higher the learning rate, the higher the weight changes. 2) during early stages oflearning and that it be increased as the net begins to converge. , links between Layers i and j). AWij is defined as AWij = -1/(8EI8wij)' ............................ (A-IS) The output, OPj' from a node in the middle (hidden) Layer j because of Pattern P is defined as The derivation that follows is similar to that for the case when the link is connected to the output layer.
F;;y' ............................. (D-I) (Fig. D-I). The model has two similar patterns, one smaller in size with longer beginning and ending positions as compared with the other. Using the scaling method AI-Kaabi and Lee proposed, we obtain scaled versions of the two patterns (Fig. D-2). We see that after scaling, the shapes have been altered considerably. Thus, if Pattern 1 was used as a training set for Model A, the neural net probably would not identify Pattern 2 as Model A.
Peterson chair In Petroleum Engineering at Texas A&M U. A. Holdltch and Assocs. _ _ __ _- - ' cal adviser at Lee AI-Kaabl Exxon Co. A. and a senior research specialist at Exxon Production Research Co. Lee holds BChE, MS, and PhD degrees In chemical engineering from Georgia Tech. He Is a member of the Career Guidance and Archie Scholarship committees and the Emerging/Peripheral Technical Committee. Lee Is a Distinguished Service Award winner and a member of the National Academy of EngineerIng. Azlz U.