Trend analysis

Trend analysis

Trend analysis in Do You Know is a way to examine how things change over time. It’s like checking the history of something to see if there are any patterns or changes that might happen again in the future. Predictive trend analysis helps to predict what might happen next based on what happened before.

Steps to perform trend analysis in Lumenore:

Step 1: After accessing “Do You Know,” select “Trend.”

Step 2: Now, click on “Create New Insight-Trend.”

To Configure Trend Analysis

Step 3: Choose a “Schema” and proceed.

Note: The schema signifies the dataset for analysis. If absent, create one (from self-service), ensuring prerequisites (A KPI, Date, and Attribute) are met.

Step 4: Select the following to configure:

• Select Trend Metric: Choose the metric for trend analysis.
• Select Trend Date Attribute & Frequency: To structure and summarize the data based on a date attribute like ship date or order date, select a frequency such as yearly, quarterly, monthly, weekly, etc.
• Trend Attribute (Optional): Specify the attribute for identifying trend patterns (e.g., Product Category, Country).
• Algorithm: You have two options: Auto-select and Manual. In Auto-select, Lumenore will choose the best machine learning algorithm for the given data. In the Manual, you can choose the algorithm you want to use for forecasting based on your preferences.

Below are the algorithms for the forecast. You can manually select them: Various time series forecasting algorithms are available, each with distinct strengths.

Linear:

• • • Used where data has linear distribution.
    • Execution time grows linearly with the size of the input data.
    • Characterized by a straight-line relationship between the input size (n) and the number of operations performed.

Logarithmic:

• • • Execution time grows logarithmically with the size of the input data.
    • Exhibit a slower and more efficient growth pattern.

Exponential:

• • • Suitable if data has exponential distribution.
    • Running time grows exponentially with the size of the input

Polynomial Regression:

• • • Regression analysis where the relationship between the independent variable (or predictor) and the dependent variable (or response) is modelled as an nth-degree polynomial.

Power:

• • • Suitable for data has power-log distribution.

• Add Filters (optional): Apply filters based on conditions.

Then click on “Next.”

Step 5: Name the insight for future reference (default suggestion provided). Then click on “Save.”

Note: If you wish to apply a filter, a window for creating filters will appear. As shown in the forecast, establish filters by groups or conditions as needed.

Step 6: Click “Execute” to initiate the analysis.

Upon initiation of execution, the system will undergo four background processes. You can also terminate the execution at any point before its completion.

This insight was crafted using quantity (trend metric), Order data- monthly (trend date attribute), and category (trend attribute) for illustrative purposes.

Now, you can see the “Trend Analysis”.