Difference between revisions of "ANLY482 AY2017-18T2 Group32: Project Overview/Methodology"
Line 8: | Line 8: | ||
| style="padding:0.2em; font-size:130%; background-color:#000000; text-align:center; " width="5%" | | | style="padding:0.2em; font-size:130%; background-color:#000000; text-align:center; " width="5%" | | ||
− | [[ANLY482_AY2017- | + | [[ANLY482_AY2017-18T2_Group_32| <font face ="Avenir Next" color="#ffffff" >Home</font>]] |
| style="background:none;" width="1%" | | | style="background:none;" width="1%" | |
Revision as of 00:05, 15 January 2018
Description | Data | Methodology |
Before performing any data analysis, it is critical to perform exploratory data analysis to understand the data better. Subsequently, we would have to handle missing data, create dummy variables for categorical features and check for correlated features.
Afterwards, feature selection methods such as ANOVA and recursive feature selection will be used to determine if the given variables are enough for us to pinpoint what has the strongest influence on the final customer review. Then, dimensionality reduction techniques such as Principal Component Analysis would be utilized to see if there is a need to create a combination of variables. Lastly, we would explore the use of Natural Language Processing and deep learning to increase personalization of the customer experience for Shopee.