Definition
Supervised Machine Learning
“Supervised learning is the use of an algorithm that uses labeled data to produce a training data set, from which the algorithm can learn”
“______ does not use labeled data or a training data set. _____ solutions identify data patterns by discovery, without training”
Unsupervised Machine Learning
AI is one application of Machine Learning
False -> Machine learning is one of the applications of AI
“Compare and contrast supervised and unsupervised machine learning.”
Supervised learning is about prediction with labeled data, while unsupervised learning aims to uncover patterns in unlabeled data
Define Business Intelligence
“Business intelligence is the use of tools (data mining, machine learning, and visualization) to convert data into actionable insights and recommendations.”
Reinforced Learning
The use of feedback loops to reward correct predictions, and punish mistakes
Hierarchically structured process that can leverage layers of machine learning, for which the output of one layer becomes the input for the next.
Deep learning
In Machine Learning most of the algorithms are unsupervised
True
“Compare and contrast machine learning and artificial intelligence.”
“Machine learning is an application of AI. In other words, there are many different applications of AI, of which machine learning is one.”
“_____ is the use of charts and graphs to represent data. Analysts use many different charts to represent data. The key is knowing which chart to use when.”
Data Visualization
Testing Data
“The testing data set must have values for the independent variables with which it will perform its analysis and the dependent variable, which the model will compare to its result. ”
“A ________ contains data for the independent variables, which influence the dependent variable, as well as correct results for the dependent variable.”
Training Data
Developers normally use 20-30% of a data set as the training data set
False -> Developers normally use 70-80% of a data set as the training data set
Compare and contrast data classification and data clustering
Data clustering is cluster data into different groups based on their different attributes. Data classification is assigning data to matching groups.
Data Association
It is the process of identifying key relationships between variables.
Classification
A supervised machine learning solution that assigns data items to specific categorie.
________ is the assignment of data items to a related group. ___ uses unsupervised learning
Clustering
“When choosing a data set, the key is to have a larger number of independent variables”
False ->
“When choosing a data set, the key is not to have a larger number of independent variables (many may not correlate to the dependent variable), but rather, to have a larger number of data records for the key dependent variables.”
Compare and contrast training data and testing data.
Training data is used to teach the model, while testing data is used to evaluate its performance
What are the four factors that affect the quality of a set of data?
Accuracy, Consistency, Conformity, Completeness