Supplier & Product Rating based on Customer Feedback : HANA Sentiment analysis
Introduction
With this quick blog I am explaining how to implement a supplier rating application using the Text Analysis capabilities of SAP HANA. I am using Sentiment analysis feature to rate a particular product and suppliers based on the customer feedback on various web sites .
Assumptions
We have the customer feedback collected from suppliers and arranged in the following format. (data is copied from multiple online suppliers. Product names and supplier names are masked )
Modelling your source Table
Now we need to load the data into SAP HANA. For that, We are creating a table of the same structure .
After once the data is loaded into the source table, we are creating a full text index on the table for the FeedbackText column using the standard configuration “EXTRACTION_CORE_VOICEOFCUSTOMER“.
Voice of the customer content includes a set of entity types and rules that address requirements for extracting customer sentiments and requests. You can use this content to retrieve specific information about your customers’ needs and perceptions when processing and analyzing text. The configuration involves complex linguistic analysis and pattern matching that includes processing parts of speech, syntactic patterns, negation, and so on, to identify the patterns to be extracted.
Once the index is created , one new columnar table will be created in the same schema where the source table is created. This table will be having the tokens as well as the corresponding sentiments values .
Now generating some visualizations on the generated sentiment data.
SAP Lumira For Reporting
Now we have the sentiment data available in our table. That means, we have the product details, Supplier details, and the type of feedback given by the customers . Now we will generate dashboard(Real time dashboards on LUMIRA if we integrate it with live feedback data on multiple web sites) .
And I love this part š
(This blog is created for DataGeek ).
Sreehari
Hi Sreehari,
Very helpful blog! I am seeking for a logic forĀ the calculation viewĀ in which TA_TYPE will work as a measure having couple of values likeĀ PositiveĀ Sentiment and Negative Sentiment based on which TA_TOKEN will show values in the graph.
TIA
Kind Regards,
Ishan Mukewar
Thanks. can you post a sample output in table as expected. I am too lazy to read my blog again. Its been a long time š
Sreehari