BACKGROUND
Electronic cigarette (e-cigarette) products are increasingly being used by youth and this trend is linked in part to pro-e-cigarette content on social media. YouTube is a popular platform used by youth to consume, produce, and share e-cigarette videos. YouTube has also become a valuable resource for researchers to learn about e-cigarette use, trends, marketing, and perceptions. However, there is currently a lack of understanding on whether or how similar e-cigarette-related search items used to obtain YouTube videos are related or are relatively mutually exclusive. This study uses novel methods to evaluate the relationship of e-cigarette-related search items to each other.
OBJECTIVE
To apply network modeling and rule-based classification to characterize the relationships between e-cigarette-related search items. We also gauged the level of importance of each search item as part of an e-cigarette information network.
METHODS
We used 16 fictitious YouTube profiles to retrieve 4201 distinct videos from 18 keywords related to e-cigarettes. We used network modeling to represent the relationships between the search items. Moreover, we developed a rule-based classification approach to classify videos. We used betweenness centrality (BC) and correlations between nodes (ie, search items) to help us gain knowledge of the underlying structure of the information network.
RESULTS
Modeling search items and videos as a network, we observed that broad search items such as “e-cig” had the most connections to other search items and specific search items such as “cigalike” had the least connections. Search items with similar words (eg, “vape” and “vaping”) and also search items with similar meaning (eg, “e-liquid” and “e-juice”) yielded a high degree of connectedness. We also found that each node has 18 connections (common videos) on average. BC indicates that general search items such as “electronic cigarette” and “vaping” have high importance in the network (0.00836). Our rule-based classification sorted videos into four categories (ie, e-cigarette devices (34-57%), cannabis vaping (16-28%), e-liquid (14-37 %), and “other” ( 8-22%)).
CONCLUSIONS
Our methods evaluated the search items in items of their importance in the network of e-cigarette-related content and visualized the relationships as well as their strength through correlations between search items. Our methods can be used to identify the importance and overlap and uniqueness of e-cigarette-related search items.
Mobile APP User Attribute Prediction by Heterogeneous Information Network Modeling