Zeinab Rika

Data Science – Manchester UK

March 1, 2020 By Zeinab Rika

Maldives Hotel reviews by sentiment analysis and opinion mining

Online retailers and businesses use text mining (opinion and sentiment analysis) for understanding customers’ needs and improve customer service. In the e-commerce industry using sentiment analysis tools can help them to become closer to the key stakeholders. Nowadays because of social media in the e-commerce industry, there is competition. Every product, brand, and the place has many reviews, so it even difficult for the customer making a decision as well as businesses to manage customer opinions. Therefore, operation and classifies words by sentiment analysis help e-commerce improve business and help customers to choose a better option.

Obviously working with text could be more difficult because it is not just about the data and needs time to prepare the dataset and then get the result and analysis results to take time to understand the problem. Maldives hotel reviews dataset has 8 attributes and 21071 instances. In this dataset, using reviews of 106 hotels in the Maldives. I used column reviews and hotel names for sentiment analysis, I used the lexicon method to count the number of positive and negative words for each hotel.

 

variables name and type in the Maldives hotel reviews dataset (21071 instances)

Variable name Data type Description
Review id Nominal The unique ID: include 9-digit integral number with two letters
Hotel Name Nominal Hotel name in the Maldives
Total review count Numeric Total number of reviews for each hotel
Review_viaMobile Nominal Customer use of mobile for review or not
Review Date Numeric Review Date: the year and date of reviews for each hotel
Guest Location Nominal Country name and city for each customer
Review Heading Text Title of review
Review Text Text review for each hotel

https://github.com/zeinabrika/text_mining_R