We then reasoned very quickly by analogy, i.e. You could also say that we collected data describing the fall of the glass. So what were we doing in our heads when we made the prediction for the state of the glass after the impact? We measured the characteristics of this event. We can watch as many tosses of a coin as we like we will never manage to recognise the necessary factors fast enough and extrapolate them accordingly in the case of a random throw. However, it is just that we cannot do with the toss of a coin. In the case of the falling glass, we quickly recognised the most important characteristics such as the material, falling height and nature of the ground and can already estimate, in the shortest time, the probability of the glass breaking by analogy reasoning from similar experiences. The reason for our inability to describe and extrapolate the processes precisely is rather down to the fact that we are not able to recognise or measure the necessary influence factors or correlate these. We shall not go into the interesting yet nonetheless rather philosophical discussions on this topic, but we are putting forward the following thesis: The vast majority of processes in our perceptible environment are not a result of coincidences. So how do we humans know what exactly will happen next in some cases and in other cases, for example that of the toss of a coin, what will not? The most frequent explanation used by laymen in this case is the description of the one scenario as “coincidental” and the other as “non-coincidental”. For example, the energy of the impact is transferred to the ground better in this case. For what it’s worth, the glass not breaking would be just as non-coincidental, since this result also follows physical laws.

Of course, the glass may stay intact “by chance” in individual cases, but this is not likely.

How are we able to achieve this rather amazing feat? We have never seen the glass which is falling in this instant break before and the physical formulas that describe the breakage of glass are a complete mystery for most of us at least. We even know in the fractions of the second when the glass is falling: There will be broken glass. We shall now look at another scenario, only this time we can predict the outcome of the situation: A glass will break if it falls from a certain height onto a certain type of ground. The physical formulas for this are all known in any case. If all influence factors such as the throwing speed and rotation angle, material properties of the coin and those of the ground, mass distributions and even the strength and direction of the wind were all known exactly, then we would be quite able, with some time and effort, to predict the result of such a coin toss. That may be correct, but the fact that nobody can predict it does in no way mean that it is impossible in principle. “Ah” you may think, “but that is a random event and nobody can predict which side of the coin will be showing after it is tossed”.

Let us take for example what is probably the simplest random event you could imagine, i.e. ĭo you believe in coincidence? As an analyst you will probably learn to answer this question in the negative or even do so already.

Work out the checksum of the result, i.e.

Whether you are already an experienced data mining expert or not, this chapter is worth reading in order for you to know and have a command of the terms used both here and in RapidMiner.ġ.1 Coincidence or not? Before we get properly started, let us try a small experiment: In this chapter we would like to give you a small incentive for using data mining and at the same time also give you an introduction to the most important terms. 5.3.1 Propagating Meta Data from the Repository and the Process. 5.2.2 Importing Data and Objects into the Repository 5.2.3 Access to and Administration of the Repository. 5.2.1 Processes and Relative Repository Descriptions.

ĥ Repository 5.1 The RapidMiner Studio Repository. 4.1.1 Sources for Displaying Results 4.2 About Data Copies and Views. Ĥ Data and Result Visualization 4.1 Result Visualization. 3.3.1 Transforming Meta Data Executing Processes. ģ Design of Analysis Processes 53 3.1 Creating a New Process. 2.3.4 Further Options of the Process View 2.3.5 Parameters View. Ģ First steps 2.1 Installation and First Repository. 1.2.1 Attributes and Target Attributes 1.2.2 Concepts and Examples. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by means electronic, mechanical, photocopying, or otherwise, without prior written permission of RapidMiner.ġ Fundamental Terms 1.1 Coincidence or not?.

Rapidminer studio association