Early-production stage prediction of movies success using K-fold hybrid deep ensemble learning model

Sahu, Sandipan; Kumar, Raghvendra; Long, Hoàng Việt; Shafi, Pathan Mohd

doi:10.1007/s11042-022-13448-0

Cited by 5 publications

(1 citation statement)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most studies conducted in cinema have focused on predicting the success and profitability of movies through different data mining or machine learning methods. [9][10][11][12][13][14][15]. Among the scarce studies that have directed their attention toward the content of cinematic works, one has considered references to a particular movie as the basis for generating a graph.…”

Section: Literature Reviewmentioning

confidence: 99%

In the Arena of the Content War: A Social Network Analysis Approach for Content Differentiation in VOD Platforms

Noroozian,

Amiri,

Ramezani

2024

IEEE Access

View full text Add to dashboard Cite

The quest for content differentiation and audience engagement is paramount in the competitive landscape of Video-on-Demand (VOD) platforms. This paper proposes a system that could revolutionize how VOD platforms approach content creation. In this context, content differentiation refers to creating unique and distinct content that stands out from the competition. By investigating VOD platform series data, the system can recommend actors and themes to VOD vendors, helping them create innovative storylines. To achieve this, we analyzed the storylines of 72 VOD movie series to extract keywords. Using these keywords, we generated a network using Social Network Analysis. By examining the network, we identified content gaps. After semantic clustering, we utilized TOPSIS to determine keywords with high centrality scores. These are a foundation for connecting peripheral keywords and generating new content. The paper's findings suggest that the proposed content-based recommender system can help VOD platforms create innovative storylines by leveraging structural holes in a TV series content keyword graph. The study also suggests successful directors take risks by combining keywords from various network parts. This approach allows them to attract a broader range of audiences and increase their chances of success.

show abstract