From f7561d22da818ef3f44568f4087a63697c811212 Mon Sep 17 00:00:00 2001 From: Kathirmani Sukumar Date: Sun, 24 Dec 2017 13:53:40 +0530 Subject: [PATCH] guidelines data visualization --- Guidelines_data_vis.txt | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) create mode 100644 Guidelines_data_vis.txt diff --git a/Guidelines_data_vis.txt b/Guidelines_data_vis.txt new file mode 100644 index 0000000..ffe7b18 --- /dev/null +++ b/Guidelines_data_vis.txt @@ -0,0 +1,25 @@ + + +1. Column types - Categories/Dimension/Groups/Qualitative , Numerics/Metric/Quantitative, Dates, Location, Text +2. Granularity of data +3. Derive new metrics + - Numerical column: Bin it + - Text - No. of characters, words, sentences, hashtags, frequency of particular word + - Date - Year, month, day, hour, second, evening/morning, weekday/holiday, Seasons + - Location- Latitude, longitude, State, Country, Avg temperature, Population, GDP +4. Univariate Analysis + - Dimension - Frequenct Count - Bar chart + - Numerical - Distribution + - Histogram, Boxplot (Outliers, skewness, distribution) + - Text - Create word cloud to understand what are frequently used words + - Date - Consider it is a dimension +5. Bivariate Analysis + - 1 Dimension + 1 Metric - Barchart, x-axis(Dimension), y-axis(Metric), Summarize(Min, max, mean, median, Sum) + - 2 Dimension: Heat Map (Color represents the frequency) + - 2 Metric: Scatterplot (Outliers, Visual Grouping, Trend, Correlation) + +6. Multivariate Analysis + - 1 Dimension & 2 Metrics - Treemap (Size is metric, color is another metric, boxes for dimension) + - Colored bar chart(bars for dimension, yaxis-metric, color-metric) + - 2 Dimension & 2 Metrics - Scatter plot (Xaxis-metric, yaxis-metric, color-dimension, symbol-dimension) + - Many metrics - Scatter plot (xaxis, yaxis, color, size) \ No newline at end of file