High Level Project Management – Data Science

Data Cleansing / Data Preparation

• Outlier Analysis / Treatment – 3R (Rectify, Retain, Remove)
• Missingness of data – Imputation – Mean, Median, Mode, Regression, KNN
• Standardization (X-Min(X)/Range(X) / Normalization (X-Mu/Sigma)) – Unitless and Scale Free
• Discretization / Binning / Grouping
• Transformation (log, exp, etc.)
  • Non-linear
  • Non-normal
  • Heteroscedasticity – unequal variance
  • Collinearity
• Dummy variable creation – One hot encoding

Exploratory Data Analysis

Earn yourself a promising career in data science by enrolling in the Data Science Classes in Pune offered by 360DigiTMG.

• First-moment business decision / Measures of central tendency
  • Mean, Median, Mode
• Second-moment business decision / Measures of dispersion
  • Variance, Standard Deviation, Range
• Third-moment business decision – Skewness
• Fourth-moment business decision – Kurtosis
• Graphical Representation
  • Univariate
    • Box Plot
      • Primary purpose – Identify outliers
      • Secondary purpose – Identify shape of distribution
    • Histogram
      • Primary purpose – Identify Shape of distribution
      • Secondary purpose – Identify outliers
    • Q-Q plot – Data are normal or not
  • Bivariate
    • Scatter plot
      • Primary purposes
        • Direction-Positive, Negative, no correlation
        • Strength – Strong, moderate, weak – Subjective; Objective – correlation coefficient; r: -1 to +1; |r| > 0.85; |r| < 0.4
        • Linear or Non-linear / Curvilinear
      • Secondary purposes
        • Scatter plot
          • Primary purposes
            • Clusters
            • Outliers
        • Feature Engineering / Feature Extraction – Using your given variables, try to apply domain knowledge to come up with more meaningful derived variables
        • Feature Selection -> Decision Tree (Information Gain), Random Forest (Variable Importance plot), Hypothesis testing, Lasso regression, Ridge regression

Supervised Learning / Machine Learning / Predictive Modelling (Y known)

• Regression Analysis (Interpret the parameters)
  • Y= Continuous -> Linear Regression
  • Y = Discrete (2 categories) -> Logistic Regression
  • Y = Discrete (> 2 categories) -> Multinomial / Ordinal Regression
  • Y = Count -> Poisson / Negative Binomial Regression
  • Excessive Zero – ZIP / ZINB / Hurdle
• KNN
• Black Box Techniques (No interpretation exists)
  • Neural Networks
  • SVM
• Ensemble Techniques
  • Stacking
  • Bagging(Random Forest)
  • Boosting (Decision Tree)

Looking forward to becoming a Data Scientist? Check out the Data Science Course and get certified today.

Unsupervised Learning (Y unknown)

• Clustering / Segmentation – Reduce the rows
  • K-Means / non-hierarchical – Upfront determine the # of clusters – Scree plot / Elbow curve
  • Hierarchical / Agglomerative – Dendrogram
  • DBSCAN
  • OPTICS
  • CLARA
  • K-medians / K-Medoids / K-modes
• Dimension Reduction – Reduce the columns
  • PCA, Factor Analysis
  • SVD
• Association Rules / Market Basket Analysis / Affinity Analysis
  • Support
  • Confidence
  • Lift Ratio > 1 => Antecedent and Consequent have strong association
• Recommender Systems
• Network Analytics
  • Degree
  • Closeness
  • Betweenness
  • Eigenvector
  • Page Rank
• Text Mining & NLP
  • BoW
  • TDM / DTM
  • TF / TFIDF

Also, check this Data Science Institute in Bangalore to start a career in Data Science.

Watch Free Videos on Youtube

Forecasting / Time Series

• Model-Based Approaches
  • Trend
    • Linear
    • Exponential
    • Quadratic
  • Seasonality
    • Additive
    • Multiplicative
• Data-Based Approaches
  • AR
  • MA
  • ES
    • SES
    • Holts
    • HoltWinters