TikTok: Analyzing User Reviews and Rating Distribution
Introduction
TikTok, a social media app owned by the Chinese company ByteDance Ltd, has gained immense
popularity since its international launch in 2017. With its focus on short-form user videos and
diverse content genres, TikTok has become a platform for creative expression, entertainment, and
social interaction across different age groups. In this essay, we will analyze key words and rating
distribution in user reviews of TikTok, aiming to gain insights into users' opinions and
experiences.
Dataset
The dataset used for this analysis consists of user reviews and comments for the TikTok app
available on the Google Play Store. The dataset provides information such as the user's name,
review score, review content, number of thumbs up received, app version, and review creation
date. The dataset contains a total of 375,151 reviews and was last updated in November 2021.This
dataset can be accessed and downloaded from the Kaggle platform using the following link:
https://www.kaggle.com/datasets/shivkumarganesh/tiktok-google-play-store-review.
Analyzing User Reviews
To begin the analysis, we performed some descriptive statistics on the dataset. The majority of
reviews, 73.6% (338,894), gave TikTok a score of 5 stars, indicating a positive rating. The
remaining reviews were distributed among other rating scores: 6.0% (27,665) with 4 stars, 4.0%
(18,620) with 3 stars, 2.9% (13,180) with 2 stars, and 13.5% (61,928) with 1 star.
We visualized the rating distribution using a bar chart, which clearly showed the dominance of
5-star ratings. This suggests that TikTok has generally been well-received by users, with a smaller
proportion expressing more mixed or negative sentiments.
Data Preprocessing
To prepare the review content for further analysis, we performed several data preprocessing steps.
First, we converted all text to lowercase to standardize the text and avoid issues with
capitalization. Next, we removed non-letter characters and spaces, ensuring that only words
remained in the text. We then split the text into individual words and removed empty strings and
common stop words. These steps helped to focus the analysis on meaningful words and improve
the accuracy of subsequent analyses.
Common Words in User Reviews
After preprocessing the review content, we analyzed the most common words used in the user
reviews. The analysis revealed that the most frequently occurring words were "app" and "TikTok."
This indicates that many users mentioned these terms in their reviews, suggesting a focus on the
app itself and its performance.
