Update README.md

README.md

@@ -118,12 +118,24 @@ The dataset was compiled from https://huggingface.co/datasets/duxprajapati/sympt
 which was then processed in terms of data-labeling using Smabbler's QueryLab platform ensuring a accurate representation of data-labels for common and rare diseases.
 
 
-Pre-processing:
+### Pre-processing:
 
-Data was pre-processed to ensure consistency and accuracy. This involved cleaning the data, handling missing values, and normalizing the binary encoding.
+The pre-processing stage is very crucial to the building of an accurate machine learning model and in terms of ensuring its reliability to be used in medical domain. 
+It involves data cleaning process which is a bit labor-intensive involving extensive manual checks for consistency and iterative validation for retaining high quality of final dataset. 
+These processes are particularly complex while dealing with medical data.
+
+Here the data was pre-processed to ensure consistency and accuracy. This involved cleaning the data, handling missing values, and normalizing the binary encoding.
 Each symptom was converted into a binary feature (0 or 1), indicating its absence or presence respectively.
 The labels were mapped to specific diseases using a detailed mapping file to ensure accurate representation.
 
+Smabbler made the pre-processing method easy by providing automated labeling,reducing the manual effort, ensuring consistency, 
+and maintained high accuracy in the pre-processed dataset,
+making it a crucial asset in building a reliable disease diagnostic model. 
+
+The data cleaning process, which would have been labor-intensive and time-consuming, was significantly expedited by Smabbler's tools and features.The platform's automation,
+standardization, and validation capabilities ensured that the pre-processing was not only quicker but also more reliable and accurate.
+
+
 Label Mapping:
 
 The labels in the dataset correspond to various diseases. A mapping file (mapping.json) was used to translate encoded labels to human-readable disease names.