Skip to content

Latest commit

 

History

History
26 lines (17 loc) · 1.95 KB

File metadata and controls

26 lines (17 loc) · 1.95 KB

News-Analysis-Using-Microsoft-Fabric

🚀 Azure Data Engineering project: News Analysis 📈

In this project, I have created an end to end solution for analyzing the latest news data. I have used the microsoft fabric for all the tools. Here's a breakdown of the journey:

workflow

1️⃣ Data Ingestion: Created pipeline in data Factory which connects to Bing API and ingest all the latest news articles as a raw json structure to the lakehouse. Source Code = https://github.com/ADNANVL/Data-Analytics-Using-Microsoft-Fabric-/tree/main/Data%20Ingestion

2️⃣ Synapse Data Engineering: Used synapse data engineering component to read the ingested raw json file and process it to a clean and structured Delta table and load that into the same Lakehouse database. Spark Code = https://github.com/ADNANVL/Data-Analytics-Using-Microsoft-Fabric-/blob/main/Data%20Transformation/process_news_data.ipynb

3️⃣ Synapse Data Science: used this to read this clean Delta table and sentiment analysis is performed further by using a description column which contains information about the news articles so basically we use this information and predicted the sentiment of the news using a pre-trained synapse machine learning model and the data is stored as a Delta table in the Lakehouse. Spark Code = https://github.com/ADNANVL/Data-Analytics-Using-Microsoft-Fabric-/blob/main/Sentiment%20Analysis/News_Sentimental_Analysis.ipynb

4️⃣ PowerBI Dashboard - Created two page report one page was autocreated by powerbi and then I created a new page which is our main news dashboard based on requirements in this dashboard I have configuration in a way that every time when I open this report only the latest news that are published in the last 24 hours will be displayed. Dashboard = https://github.com/ADNANVL/Data-Analytics-Using-Microsoft-Fabric-/tree/main/PowerBI%20Report