Clarifying Data Analytics Concepts for I. E.

Clarifying Data Analytics Concepts for I. E.

1. Data come from different areas, such as product process design, assembly, material planning, quality control etc...

2. Data analytics wants to analyse raw data in order to discover hidden patterns and relationships among different variables.

3. Background Main Concepts

   • It focus on statistics and how some of its algorithms are used to investigate datasets of huge dimensions, they are called big data problems
     • Data mining is the extraction of implicit. The idea is to construct numerical algorithms that investigate datasets to extract hidden patterns and regularities.
       • Machine learning is the field that developed most of the algorithms to find and describe structural patterns in data.
         • Data analytics is a trend word, since it is strictly connected with the buzzword "Big Data". Data analytics is composed both of data minning and statistical models to clean data at the begginign and to validate rules at the end.
           • It is clear big data refer to a type of datasets, in which variables are different in type and nature, for this reason, the compel to be analysed with a number of specific techniques, models and algorithms, that belong to data analytics.

4. Extraction of value from data

   • Last decade manufacturing has experienced a tremendous shift toward digitalization.
   • The extraction of value from data is a new challenge that companies are now experiencing
     • Manufacturing companies need to embrace data analytics to remain competitive in the global market.
   • Through data minning it is possible to achieve competitive advantages and the superiority of data minning and machine learning over the experimental techniques lies in the fact that data can be collected during the normal operations of the manufacturing process.

5. Machine learning techniques in smart manufacturing

   • Linear methods for regression were largely developed in the pre-computer age of statistics.
     • Linear method for classification uses the variable´s characteristics to identify which class the variable belong to.
       • Decision tree based method approach to the problem of learning from a set of independt instances leads naturally to a style of representation called a decision tree.
         • Neural Networks are to extract linear combinations of the inputs as few features, and then model the traget output as a nonlinear function of tthese features.
   • Cluster Analysis implements the grouping or segmantation of different objects into subsets or clusters as primary goal.
     • SOM is a method that can be viewed as a constrained version of kmeans clustering, in which the original high-dimensional observations can be mapped down into the two-dimensional coordinate system