Introduction
Recommendation in machine learning is a technique used to predict and suggest relevant items, products, or content to users based on their preferences, historical behavior, and similarities with other users. It is widely used in various applications, such as personalized product recommendations on e-commerce websites, movie recommendations on streaming platforms, and content suggestions on social media platforms. Recommendation systems play a crucial role in enhancing user experiences, driving engagement, and increasing business revenue. Example When browsing a music streaming platform, the recommendation system suggests new songs and artists based on your listening history and preferences, introducing you to music that aligns with your taste. Now, we will see how SAP PaPM provides solutions to businesses with an example.
Model table has been created with two fields: User ID, which contains user details, and Movie Name
In the Environment fields, master data is maintained for User ID and Movie Name.
User ID Master Data
Movie Name Master Data
The data mentioned below has been uploaded into the Model table, which contains information about which users watched which movies.
For better understanding, let me explain the uploaded data. It contains information about which users watched which movies.
Hope above mentioned capture will be clear how the data is uploaded. Now we will see how to handle this in SAP PaPM.
Create the fields as per below mentioned capture. Theas fields will be used in output fields in the machine learning function.
Create the Machine learning function and assign the input function (Model Table)
Assign the fields which is created in the environment.
Create the rule with rule type ‘Recommendation’
Now assign the fields.
Input Fields
User ID: User ID created which is created in the model table
Item Fields: Movie Name which is created in the model table
Minimum Support: Default value is 2 as per SAP document. For more information refer help.SAP.com
Minimum Confidence: The default value is 0.5.as per SAP document. For more information refer help.SAP.com
Output Fields – Assign the fields from the signature tab.
Activate and Run. Now the system recommending some movies for some user based on the similar movie watched.
Now, let’s delve into an explanation of how the system generates recommendations. User 1001 has watched 6 movies according to the data uploaded in the model table. This user has only watched movies with higher numbers compared to other users. However, the system is not recommending any movies.
User 1002 has watched 5 movies. When compared with other users, only ‘1001’ has watched movie M6. As a result, the system does not make recommendations due to the utilization of a minimum support of 2, along with a confidence level of 50%.
User 1003 has watched 4 movies. When compared with Users 1001 and 1002, movie M5 is a common movie that has not been watched by User 1003. As a result, the system is recommending the movie M5 with a score of ‘0.66’.
User 1004 has watched 3 movies as per the data uploaded in the model table. When compared with Users 1001, 1002, and 1003, movies M4 and M5 are common movies that User 1004 has not watched. As a result, the system is recommending movies M4 and M5 with scores.
Same rule as per User 1004.
In conclusion, SAP PaPM machine learning capabilities offer tailored recommendations, optimizing user engagement and driving business success
Thank you for reading! Please feel free to leave any comments or feedback. SAP PaPM
Similarly, to explore more of our blogs click here.
Moreover, you can check our video blog also – Trijotech
Disclaimer: All the opinions are solely for information purposes and the author doesn’t recommend or reject any tools. It should be done after your own due diligence.
