دانلود کتاب داده کاوی

Artificial intelligence (AI) has been on everyone’s lips for quite some time now. Generative AI in particular is constantly opening up new areas of application. Generative AI is based on machine learning processes, which brings us to data mining: data mining encompasses a range of basic technologies for analyzing mass data, which in turn are relevant for training AI. We hear or read buzzwords such as big data, predictive analytics, data science, or machine learning almost every day. Data is collected in many areas, such as research, to discover new correlations or improve existing processes, while companies collect data to achieve higher sales figures through individualized marketing or to reduce machine or plant downtime. Even in medicine, data is collected and evaluated to enable better diagnoses and therapies. With the help of such large data collections, methods for text or character recognition, speech comprehension, and image recognition have been developed that are already integrated into many systems today and are giving rise to intelligent applications. Data is collected for various reasons: first of all, simply because collecting and storing data is technically and financially possible. Secondly, there are legal requirements that data must be stored, at least for a certain period of time. Thirdly, and here we are getting closer to data mining, collecting data is also carried out for a specific purpose, i.e. a benefit is expected. What data is collected? All data, unless collection is r estricted by law: technical data, business data, and even private data. In this book, we examine techniques that can be used to evaluate large amounts of data. We focus on the processes and techniques for analyzing structured data that are commonly associated with data mining. Specifically, our book covers the following topics: -First, we provide an overview of data mining and discuss basic concepts and procedures. -Next, we describe application classes in order to identify the possible uses of data mining. -Models are derived from the data. We show how these models can be represented or stored. -The methods for analyzing the data, data mining in the narrower sense, naturally take up most of the space in this book. -The data must be prepared to enable analysis and improve data quality. Data quality can have a significant impact on results. A separate chapter is therefore devoted to data preparation. -This is followed by an evaluation of the analysis results: Are they new, are they really relevant? This chapter also considers possible visualizations of the results as a form of evaluation. A suitable graphical representation helps both to assess the quality of the results and to build confidence in a model. Finally, we run through data mining using a concrete example. We thus illustrate the data mining process and apply the analysis techniques presented in the book. Statistical methods for data evaluation are only touched upon briefly. The areas of text analysis, known as text mining, as well as web mining, are only outlined. Unlike many other books, we separate the models, such as decision trees or artificial neural networks, from the data mining methods that use them, such as the generation of a decision tree for classification. This makes it clearer that some models can be used for different tasks.