Lesson 6

• Data, as we all know, is available in its basic form as information. Before it becomes significant data, the raw piece of information must go through its own journey.
• These bits of data might be difficult to comprehend in their raw form and cannot be fed directly into algorithms.
• The two most crucial levels on this progression are data analysis and data interpretation.
• Data analysis and data interpretation are completely separate and follow a certain order in the life cycle of data science.
• Data Analysis


  Studying the organized material in order to discover inherent facts. The data are studied from as many angles as possible to explore the new facts.
• Data Analysis


  The process of bringing order, structure, and meaning to the collected data.
• Data Analysis


  A methodical approach to apply statistical techniques for describing, exhibiting, and evaluating the data. It helps in driving meaningful insights, form conclusions, and support the decisions making process.
• Data Analysis Tools


  • Python
  • Tableau
  • KNIME
  • enRefine
  • R
  • Power BI
  • RapidMiner
  • NodeXL
  • SAS
  • QlikView
  • Solver
  • io
  • Apache Spark
  • Microsoft Excel
  • Op
  • enRefine
• Purposes of Data Analysis

  • Description
  • Construction of Measurement Scale
  • Generating empirical relationships
  • Explanation and prediction
• Nominal Scale


  The number serves as nothing more than labels.
• Ordinal Scale


  Such numbers are used to designate an ordering along some dimensions such as from less to more, from small to large, from sooner to later.
• Interval Scale


  The interval provides more précised information than ordinal one. By this type of measurement, the researcher can make exact and meaningful decisions.
• Ratio Scale


  It has two unique characteristics: the intervals between points can be demonstrated to be precisely the same and the scale has a conceptually meaningful zero point.
• Functions of Data Analysis

  • Examining the statement of the problem
  • Examining each hypothesis of the problem
  • Studying the original records of the data before data analysis
  • Analyzing the data for thinking about the research problem in lay man's term
  • Analyzing the data by attacking it through statistical calculations
  • Thinking in terms of significant tables that the available data permits for the analysis of data
• Descriptive Statistics


  • Measures of Central Tendency
  • Measures of Variability
  • Measures of Relative Position
  • Measures of Relationship
• Inferential Statistics


  • Significance of Difference between Means
  • Analysis of Variance
  • Analysis of Co-Variance
  • Correlation Methods
  • Chi Square Test
  • Regression Analysis
• Descriptive Analysis


  Describes the data, captures and summarizes the past using measures of central tendency, measures of dispersion, visualizing using dashboards.
• Diagnostic Analysis


  Helps dig further by creating detailed, informative, dynamic, and interactive dashboards to answer why something has happened. It separates the root cause of the problem and identifies the source of the patterns.
• Predictive Analysis


  Predicts the likelihood of an event, forecasting any measurable amount, risk assessment, and segmenting customers into groups. It employs probability and uses statistics and machine learning algorithms.
• Prescriptive Analysis


  Collaborates the learnings from what has happened with what is likely to happen to help with what measures to maximize the primary business metrics. It prescribes the best course of action, strategies.
• Cognitive Analysis


  Aims to mimic a human brain to perform tasks like a human does. It combines technologies such as artificial intelligence, semantics, machine learning, and deep learning algorithms.
• Prescriptive analysis


  Result-oriented, collaborates learnings from what has happened with what is likely to happen to help with what measures to maximize the primary business metrics, prescribes the best course of action and strategies
• Prescriptive analysis


  • Not predicting one individual standalone event but a collection of future events using simulation and optimization
  • Heavily applied in financial, social media, marketing, and transportation domains
  • Uses include recommending products or movies, suggesting strategies to maximize returns and minimize risk
• Cognitive analysis


  Mimicking the human brain to carry out tasks
• Cognitive analysis


  • Combines technologies such as artificial intelligence, semantics, machine learning, and deep learning algorithms
  • Learns and generates data using available data, retrieves features and hidden patterns
• Real-time data cognitive analysis applications


  • Object detection, machine translations, virtual assistants, chatbots
• Interpretation


  Adequate exposition of the true meaning of the material presented in terms of the purposes of the study and the topic involved
• Purposes of data interpretation

  • To throw light on the real significance of the material
  • To understand implications of the data
  • To provide hints of conclusions and recommendations
  • To show the values of greatest worth that has resulted from the research
  • To refer important generalization
• The researcher should not ignore factors which are unstudied or not selected for study
• The researcher should not over-interpret the expected results or exercise defense mechanism in interpreting the results
• Quantitative data interpretation


  Applicable for measurable or numerical data, using statistical modeling methods, visual depictions like charts and tables
• Quantitative data analysis methods

  • Descriptive statistics (measures of central tendency and dispersion)
  • Inferential statistics (generalizing or inferring from sample data)
• Qualitative data interpretation


  Implemented for textual and descriptive categorical data, first coded and converted into numerical data for analysis
• Types of categorical data

  • Nominal (no ranking or order)
  • Ordinal (ranked or ordered)
  • Binary (two categories)
• Importance of data analysis and interpretation

  • Informed decision-making
  • Identification of trends and forecasting needs
  • Cost-efficient
  • Providing clear insights
• Data analysis


  Precedes data interpretation