Page 162 - Data Science class 10
P. 162

IDENTIFYING PATTERNS




                                                                                    03








                    Learning Outcome



                 3.1  Facts and Opinion                      3.2.  What is Partiality, Preference and Prejudice?
                 3.3.  How to Identify the Partiality, Preference and Prejudice?
                 3.4.  Probability for Statistics            3.5.  What Is a Sampling Error?
                 3.6.  The Central Limit Theorem             3.7.  Why is the Central Limit Theorem Important?


        In our digital age, patterns are everywhere. Either a pattern may be seen visually or it can be mathematically
        observed. The classification of data is based on previously acquired knowledge or on statistical data extrapolated
        from patterns and/or their representation. The choice of attributes and representation of patterns is a crucial step.
        A good representation is one that makes use of discriminating attributes and also reduces the computational
        burden of pattern classification.


        3.1. FACTS AND OPINION

        It is very important, as an active reader, to distinguish between fact and opinion in writing. Facts are true and
        cannot be argued with, because they can be proven and are supported by evidence, while opinions vary according
        to the attitudes of the writer. But keep in mind that information might be manipulated to support the author's
        viewpoints. When people write articles, they often select facts that support their opinion.

        When you read this kind of text, you need to be able to tell the difference between facts that can be backed up by
        evidence and ‘facts’ that have been made up by the writer. You can do this by:
        1.  checking that any reports that are quoted actually exist and refer to the fact in question

        2.  finding other sources that give opposing points of view
        3.   thinking of the reason for which the text has been written and its audience (for example, an article in a popular
           newspaper may have been written to sell the newspaper rather than to give a true or fair account of an incident)
        4.  looking for sensationalist words, such as ‘horror’ or ‘disgraceful’, which might influence readers’ opinions

        An objective piece of writing is based on facts and can be backed up with evidence. Subjective writing expresses
        opinions and feelings rather than facts.


        3.2. WHAT IS PARTIALITY, PREFERENCE AND PREJUDICE?

        We often come across circumstances where if we have a special fondness towards a particular thing, we tend to
        be a little biased toward it. This, in majority cases may affect the outcome or you can say it can alter the results to
        favour particular things. Naturally, it is not the right way of dealing with the data on larger scale. It is said that most
        (almost all) big datasets generated by Machine Learning (ML) powered systems are biased.

          160   Touchpad Data Science-X
   157   158   159   160   161   162   163   164   165   166   167