Analytics & Visualization Samples for Academic Graph

Última atualização: 07/05/2019
Editar no GitHub

This project aims to help data scientists become familar with the Microsoft Academic Graph through analystics and visualization samples using Data Lake Analytics (USQL) and Power BI.

Samples

The project contains 13 samples

Getting Started

Pre-requisites

Gather the information that you need

Before you begin, you should have these items of information:

:heavy_check_mark: The name of your Azure Storage (AS) account containing MAG dataset from Get Microsoft Academic Graph on Azure storage.

:heavy_check_mark: The name of your Azure Data Lake Analytics (ADLA) service from Set up Azure Data Lake Analytics.

:heavy_check_mark: The name of your Azure Data Lake Storage (ADLS) from Set up Azure Data Lake Analytics.

:heavy_check_mark: The name of the container in your Azure Storage (AS) account containing MAG dataset.

Create database from MAG data before running analytics examples

In prerequisite Set up Azure Data Lake Analytics, you added the Azure Storage (AS) created for MAG provision as a data source for the Azure Data Lake Analytics service (ADLA). In this section, you submit an ADLA job to create database from MAG data.

  1. In the Azure portal, go to the Azure Data Lake Analytics (ADLA) service that you created, and select Overview > New Job.

  1. Copy code in samples/CreateDatabase.usql and paste into the code block.

  2. Provide a Job name and select Submit.

  1. The job should finish successfully.

Running Example Analytics

  1. Download or clone the repository.
  2. Open the solution /src/AcademicAnalytics.sln
  3. For each tutorial there should be: A USQL script(.usql), a Power BI report(.pbix), a Power BI template(.pbit) and a README explaining the tutorial.
  4. Althought each tutorial is different, running the USQL script as is and filling out the Power BI template using the same USQL parameters should give you a Power BI report with visualizations that match the Power BI report example included in the tutorial. Since the Microsoft Academic graph is contently improving, different graph verions may give you slightly different results.

Working with USQL scripts

  • How to run

    • Make sure you have selected your ADLA account

    - Build the script first to validate syntax

    - Submit your script to your ADLA account

  • How to view the results

    • You can view the results via azure portal

Using Power BI

  • Make sure USQL script finished sucessfully
  • Open up corresponding Power BI Template(.pbit) from file explorer (Visual studio doesn't recognize Power BI files)
  • Enter your ADL information and parameters corrisponding to your scripts
  • Make sure the parameters cases are the same as your script and "click" to load

Resources