Flashcards
Data Description
Cards in this deck (8)
- 1What is a dataset?A dataset is a collection of objects and attributes. E.g customer name, age, address and country. Attributes can be numerical or categorical.
- 2Data Quality-Accuracy; How accurate?
-Completeness; Any missing data?
-Consistency; Any inconsistency?
-Timeliness; Are there any timely updates?
-Believability; How trustworthy?
-Interpretability; How easy is it to understand? - 3TerminologyPopulation: A collection of things under consideration.
Sample: A proportion of a population selected for sampling.
Statistics: A summary of measure computer to describe a characteristic of the sample data. - 4Correlation CoefficientUsed to measure the degree of a linear relationship between two variables. If y>0 than positive correlation, however it y<0 its negative correlation.
- 5Mean, Medium and ModelMean: Sensitive to extreme values e.g Outliers
Medium: Middle values
Mode: Most common number in the dataset. Unimodal if one one number is the model and Bimodal for two modes - 6Scatter plots:Shows the relationship between two variables. Linear (Straight line) demonstrates a negative relationship. Non-linear curved relationship shows no relationship.
- 7Normal Distribution- 68 percent of all values are within one standard deviation on either side of the mean.
- 95 percent are between two standard deviations on either side of the mean. - 8Frequency TableContains the outcome along with its frequencies. Allow us to see better distribution of the marks.
Information
- Description
- Bachelors Degree Data Mining Flashcards on Data Description, created by Jason Dsilva on 01/05/2016.
- Created by
- Jason Dsilva
- Views
- 0
- Rating