Available via license: CC BY 4.0
Content may be subject to copyright.
Connecting the Dots: Leveraging Social Network Analysis to
Understand and Optimize Collaborative Dynamics Within the
Global Film Production Network
Mehrdad Maghsoudi, Saeid Aliakbar, Sajad HabibiPour
Abstract
In recent years, the global film industry has observed a notable surge in international cooperation
and cross-border investments. However, a comprehensive overview of these collaborative
investments within the industry is lacking. This study employs social network analysis to delve
into the possibilities that lie in collaborative efforts and joint investments within the film sector.
The research constructs a network of 150 countries based on shared creative elements in their film
productions, comprising over 7800 interconnected links. Employing measures of centrality, certain
pivotal nations such as the United States, China, and England emerge as influential nodes,
showcasing a strong potential to steer industry growth through collaborative engagement.
Through a more detailed exploration involving community identification, distinct clusters centered
around thematic commonalities that have converged through joint creative endeavors become
evident. For example, the "Global Thrill Seekers" community focuses on action films, whereas the
"Cultural-Social Cinema Group" addresses worldwide cultural and social issues. Each of these
communities presents distinctive perspectives for international cooperation and the collaborative
creation of content. This analysis significantly enhances our understanding of the global film
network's structure and dynamics, while concurrently highlighting promising pathways for future
investment and collaborative initiatives. The research underscores the critical role of leveraging
social network analysis methodologies to optimize informed decision-making concerning
collaborative investments, thereby paving the way for anticipatory outcomes. This study not only
contributes insights but also serves as a model for investigating data-centric participation within
the creative industries.
Keywords: Film Industry, Joint Ventures , Social Network Analysis (SNA) , Collaboration in
Film , Global Film Market
1. Introduction
The film industry has played a significant role in shaping and influencing various aspects of life
and the current world (Vlassis, 2016). From its inception, the film has captivated audiences and
provided a platform for storytelling, entertainment, and cultural expression(Vlassis, 2016; Vogel,
2020). Over the years, the film industry has evolved into a global phenomenon, attracting increased
investments and fostering international cooperation(Dašić & Kostadinović, 2022).
In recent years, there has been a notable surge in investment within the film industry(Dastidar &
Elliott, 2020). This trend can be attributed to several factors, including advancements in
technology, growing demand for diverse content, and the increasing globalization of the film
market(Betzler & Leuschen, 2021). As a result, stakeholders have recognized the immense
potential and profitability of the film industry, leading to a rise in financial commitments from
various entities(Behrens et al., 2021).
Furthermore, the film industry has witnessed a substantial increase in international cooperation
and investments(Guo, 2023). Filmmakers, studios, and production companies are increasingly
engaging in cross-border collaborations, joint ventures, and co-productions to leverage resources,
expand market reach, and enhance creative synergies. This trend highlights the industry's
recognition of the value in fostering global partnerships to create successful and impactful film
experiences(Yayla et al., 2023).
However, despite the growing importance of collaborations and joint ventures in the film industry,
there remains a need for comprehensive analysis, review, and forecasting in this domain.
Quantitative and data-based analyses are crucial for understanding the dynamics, outcomes, and
potential of these partnerships(Maghsoudi & Nezafati, 2023). Unfortunately, such analyses are
often lacking, limiting the industry's ability to make informed decisions and optimize collaborative
opportunities(Bulgurcu et al., 2018; Lee et al., 2020).
One powerful technique that can address this gap is social network analysis (SNA). Social network
analysis is a methodological approach that examines the relationships and interactions between
individuals, groups, organizations, or entities within a social system. By applying SNA,
researchers can uncover hidden patterns, identify influential actors, analyze information flows, and
predict outcomes within complex social, economic, and cultural networks(M. A. M. A. Kermani
et al., 2022; Zohdi et al., 2022).
The aim of this paper is to utilize social network analysis in examining and evaluating
collaboration opportunities and joint investments in the film industry. By employing SNA
techniques, our objective is to elucidate the structure and dynamics of collaborative networks in
the industry, identify key stakeholders and their roles, and provide insights for future decision-
making..
The following sections of this article will include a literature review, where we explore existing
research and theories related to collaborative ventures in the film industry. The methodology
section will outline the approach and data sources used for our social network analysis. In the
results section, we will present our findings and analysis based on the applied methodology. The
discussion section will interpret the results, explore their implications, and discuss potential
avenues for further research. Finally, the article will conclude with key insights, recommendations,
and suggestions for policymakers, industry practitioners, and researchers interested in enhancing
collaboration and predicting outcomes in the film industry.
2. Literature review
2.1. Investments in the film industry
Investments in the film industry play a vital role in the development and production of films, as
well as the overall sustainability and growth of the industry(Johnsen, 2023; Messerlin & Parc,
2018). These investments can come from various sources, including studios, production
companies, individual investors, and even crowdfunding platforms(Dastidar & Elliott, 2020).
Understanding the different types of investments in the film industry and their potential returns is
essential for both investors and industry professionals(Johnsen, 2023).
One common type of investment in the film industry is the financing of film production. Investors
provide the necessary capital to fund various aspects of filmmaking, such as pre-production,
production, post-production, marketing, and distribution. These investments can take the form of
equity financing, where investors become shareholders in the film project, or debt financing, where
investors provide loans to be repaid with interest(Johnsen, 2023).
Another type of investment in the film industry is through film funds or investment vehicles. Film
funds pool together funds from multiple investors and allocate them to various film
projects(Dastidar & Elliott, 2020). These funds are often managed by experienced professionals
who assess the potential profitability and viability of film projects. Investors in film funds benefit
from diversification, as their investments are spread across multiple projects, reducing the risk
associated with investing in a single film(Johnsen, 2023).
Investments in the film industry can yield different types of returns. One primary return on
investment is box office revenue(Dastidar & Elliott, 2020). When a film performs well at the box
office, generating substantial ticket sales, investors can receive a share of the profits. However,
box office success is not guaranteed, and films can also underperform, resulting in losses for
investors(Johnsen, 2023).
In addition to box office revenue, films generate income from various revenue streams, including
distribution deals, home video sales, streaming platforms, merchandise, and licensing. Investors
may receive a portion of these revenues, depending on the terms of their investment
agreements(Kübler et al., 2021). It is important to note that revenue-sharing models and profit
participation can vary widely, depending on the specific agreements between investors and
filmmakers(Johnsen, 2023).
Investments in the film industry also offer intangible returns, such as exposure and prestige.
Successful films can elevate the profile and reputation of investors, opening doors to future
investment opportunities and industry connections(Dastidar & Elliott, 2020). Furthermore,
investments in the film industry provide individuals with the opportunity to support and contribute
to the creation of meaningful and impactful storytelling(Johnsen, 2023).
However, it is crucial to acknowledge the inherent risks associated with film investments. The film
industry is highly competitive, and not all projects achieve commercial success. Factors such as
audience preferences, market conditions, critical reception, and marketing strategies can
significantly impact the financial performance of a film. Therefore, investors must conduct
thorough due diligence, assess the potential risks and rewards, and diversify their investment
portfolios to mitigate risks(Johnsen, 2023).
2.2. Social Network Analysis (SNA)
Social Network Analysis (SNA) is a methodological approach that examines the relationships and
interactions between individuals, groups, organizations, or entities within a social system. It
provides a powerful framework for understanding the structure and dynamics of social networks
and their influence on various phenomena. The origins of social network analysis can be traced
back to the early 20th century, with notable contributions from sociologists, anthropologists, and
mathematicians(HabibAgahi et al., 2022).
One of the pioneers of social network analysis was Jacob Moreno, who developed the sociogram
in the 1930s. Moreno used this visual representation to depict social relationships and patterns
among individuals within a group. This innovative approach laid the foundation for the formal
study of social networks and their role in shaping human behavior and social dynamics(M. A.
Kermani et al., 2022).
Social network analysis views events and phenomena through the lens of relationships and
connections(Maghsoudi, Jalilvand Khosravi, et al., 2023). It recognizes that social structures and
patterns of interaction can significantly impact the flow of information, influence, resources, and
even the spread of behaviors or ideas(HabibAgahi et al., 2022). By analyzing the structure and
properties of social networks, researchers can uncover hidden patterns, identify key actors or
nodes, analyze information flows, and predict outcomes within complex social, economic, and
cultural systems(Maghsoudi & Shumaly).
The main components of social network analysis are nodes, ties, and attributes. Nodes, also known
as actors or vertices, represent individuals, groups, or organizations within the network. Ties, also
referred to as edges or relationships, represent the connections or interactions between the nodes.
These ties can be directed (one-way) or undirected (mutual)(Jalilvand Khosravi et al., 2022).
Attributes refer to the characteristics or properties associated with nodes or ties, such as age,
gender, profession, or the strength of a relationship(M. A. M. A. Kermani et al., 2022).
Social network analysis employs various indicators to analyze and measure the structure and
properties of social networks. Three key indicators include degree, density, and centrality.
Degree is a basic measure of the number of connections or ties that a node has within a network.
It provides an indication of an individual's popularity, influence, or connectedness within the
network. The degree of a node can be calculated by counting the number of ties it has. In a directed
network, the degree can be further classified as in-degree (number of incoming ties) and out-degree
(number of outgoing ties) to capture the flow of information or influence(Maghsoudi, Shokouhyar,
Khanizadeh, et al., 2023).
Density is a measure of the interconnectedness or cohesion within a network. It quantifies the
extent to which the nodes in a network are connected to one another. Density is calculated by
dividing the actual number of ties present in the network by the total number of possible ties.
Higher density values indicate a more interconnected network, while lower density values suggest
a more fragmented or sparse network(Maghsoudi & Shumaly).
Centrality indicators measure the importance or prominence of nodes within a network. They
identify nodes that occupy central positions and have a significant impact on information flow,
influence, or communication. Some commonly used centrality indicators include degree centrality,
betweenness centrality, and closeness centrality(M. A. Kermani et al., 2022; Maghsoudi,
Shokouhyar, Ataei, et al., 2023).
• Degree centrality measures the number of ties that a node has relative to other nodes in the
network. It provides an indication of an individual's popularity or influence within the
network. Degree centrality can be calculated by dividing the number of ties connected to a
node by the total number of nodes minus one.
• Betweenness centrality captures the extent to which a node lies on the shortest paths
between other nodes in the network. It measures the node's potential to control or facilitate
the flow of information or resources between other nodes. Betweenness centrality can be
calculated by determining the fraction of shortest paths in the network that pass through a
particular node.
• Closeness centrality reflects how quickly a node can access information or resources from
other nodes in the network. It measures the average distance from a node to all other nodes
in the network. Nodes with higher closeness centrality can disseminate information or exert
influence more efficiently. Closeness centrality can be calculated by summing the shortest
path distances from a node to all other nodes and dividing it by the total number of nodes
minus one.
The concept of community in social network analysis refers to groups or subsets of nodes that are
densely connected internally and sparsely connected externally. Communities represent clusters
of nodes that exhibit stronger ties among themselves compared to nodes outside the community.
Detecting and understanding communities in a network can provide insights into the social
structure, subgroups, and the flow of information or influence within the network(Maghsoudi,
Shokouhyar, Khanizadeh, et al., 2023).
A community is a cluster of objects that exhibit closer relationships with each other than with other
objects in the dataset. Within a community, individuals interact more frequently with fellow
members than with those outside the group (Figure 1). The proximity of entities within a
community can be evaluated by analyzing their similarity or distance from one another (Bedi &
Sharma, 2016). In the context of a social network, a community can be likened to a cluster within
the network (Said et al., 2018).
Figure 1: Communities clustering (Ding et al., 2021)
Community detection in network models involves creating community structures through
clustering, as illustrated in Figure 2. According to the definition, a community is composed of
nodes that have strong associations with one another, and these connections are stronger within
the community than with nodes outside of it. Communities, representing groups of nodes with
shared and similar properties, are valuable tools for network analysts seeking to comprehend
interactions and cohesive sub-groups within the network (El-Moussaoui et al., 2019). Modularity,
as defined by Newman, serves as a measure of social network clustering performance (Newman,
2006). Nodes that are interconnected exhibit a positive correlation with modularity.
In a weighted network with n nodes, the algorithm initiates by treating each node as an individual
community. Thus, initially, there are multiple separate communities. The algorithm proceeds by
evaluating each node (i) and identifying a neighboring community (j) that, when joined, maximizes
the modularity index while removing node i from its current community. Node i is then added to
community j if this change results in an increase in modularity; otherwise, node i remains in its
original community. This process is iteratively repeated for all nodes until no further
improvements can be made, leading to a locally optimal point in the first phase. This point
represents a state where altering the communities of nodes does not yield any additional modularity
gains.
In the second phase of the algorithm, the process continues by merging smaller communities that
can be combined to form larger ones. These two phases persist until there are no more changes in
the communities, and the modularity index reaches its maximum state.
As depicted in Figure 2, after completing the first phase and finding the local optimal point, the
algorithm successfully identifies four distinct communities. In the second phase, it attempts to
merge these four communities, eventually condensing them into two larger communities to achieve
the highest possible modularity index, at which point the process comes to a halt.
Figure 2: Community detection based on increasing modularity(Maghsoudi, Jalilvand Khosravi, et al., 2023; Yap et al., 2019)
2.6. Related Works
In the "Related Works" section of this manuscript, we explore a range of studies investigating
different facets of the film industry. (Leem et al., 2023) present a 2023 study titled "Towards Data-
Driven Decision-Making in the Korean Film Industry: An XAI Model for Box Office Analysis
Using Dimension Reduction, Clustering, and Classification." This study centers on the growth of
the Korean film market and the rising significance of explainable artificial intelligence (XAI)
within the industry. To address the competitive nature of the market and the substantial expenses
tied to film production, the authors propose the DRECE framework (Dimension REduction,
Clustering, and classification for Explainable artificial intelligence). This framework employs
dimensionality reduction techniques to transform complex data into a two-dimensional format,
employs K-means clustering to group similar data points, and utilizes machine-learning models to
classify movie clusters. Notably, the integration of XAI techniques ensures transparency in
decision-making processes, offering insights that support film industry professionals in enhancing
box office performance and optimizing profits. Adoption of the DRECE model holds the potential
to provide fresh perspectives and insights, empowering decision-makers to strategically navigate
the Korean film market for successful outcomes.
Another significant contribution comes from (Juhász et al., 2020), who delve into collaboration
networks within the creative industries, with a specific focus on film production. This 2020 study
examines the dynamics between creators occupying central and peripheral positions within these
networks and how such roles impact creative success. While central creators benefit from prestige
critical for creative achievement, they may lack exposure to diverse ideas originating from the
periphery. The authors introduce the concept that central creators can enhance their prospects of
creative success by acting as intermediaries, bridging the gap between the core and peripheral
collaborators. This hypothesis is substantiated using a distinctive dataset encompassing Hungarian
feature films from 1990 to 2009, dissecting the dynamic collaboration network among movie
creators. The findings indicate that central creators who also serve as connectors between the core
and periphery are notably more likely to achieve award-winning creative outcomes.
(Ding et al., 2022) contribute to the field through their study conducted in 2022, where they present
the "WE model," a machine learning model tailored for predicting the derivatives market for films.
The authors highlight that movie industry revenues stem not just from box office sales but also
from merchandising, advertising, home entertainment, and other sources. Intriguingly,
merchandising can even surpass box office sales in profitability. However, traditional market
research techniques for forecasting merchandising markets across diverse film genres are labor-
intensive and constrained. In response, the authors propose the WE model, which amalgamates
three machine-learning algorithms to scrutinize crucial movie attributes and accurately forecast
movie merchandising markets. By discerning the interplay between film attributes and
merchandising markets, the WE model attains a 72.5% accuracy rate in predicting and assessing
the derivatives market during testing. The study concludes that machine learning holds the
potential to enable data-driven forecasting and management of movie merchandising markets,
thereby demonstrating a successful application of machine learning techniques in predicting movie
derivatives markets based on film attributes.
(Mateer, 2020) explores the landscape of academic-industry collaboration within the realm of
commercial film and television production in the 2019 study titled "Academic-Industry
Collaboration for Commercial Film and Television Production: an exploration of case studies."
The study emphasizes the burgeoning collaborative models between academia and the film and
television industry, aiming to enhance the practical relevance of academic programs while
supporting the industry in identifying and nurturing new talent and optimizing production
expenses. The paper evaluates two collaborative paradigms: the University as a "Production
Partner" and the University as a "Service Provider." Drawing from diverse international
collaborations, the study probes the structural aspects of these partnerships, the alignment of
stakeholder needs, the benefits reaped by students and graduates, and the overall efficacy of such
initiatives. A specific case study involving the University of York, UK, and Green Screen
Productions Ltd. is also examined, underscoring the significance of alignment between
engagement and institutional objectives for successful collaboration. The research posits that these
collaborative models can be adapted to various global media contexts, contingent upon careful
consideration of participants' needs and expectations.
(Ebbers & Wijnberg, 2012) contribute insights on the influence of reputation dimensions on
distributor investments within the film industry. Their research discovers that a favorable
reputation in a congruent dimension positively impacts investor behavior. Conversely, favorable
reputations in incongruent dimensions weaken this positive effect, elucidating the phenomenon
known as "reputational category spanning."
(Parc, 2018) investigates the repercussions of protectionist policies on the success of the Korean
film industry. The study's findings indicate that protectionist measures such as import quotas and
subsidies play a limited role in bolstering the industry's prosperity. Instead, market-friendly
conditions and business activities emerge as pivotal factors in enhancing the industry's
competitiveness.
(Powers, 2015) delves into the anticipated returns on tangible investments within the film industry.
The study underscores the positive correlation between expected returns and the idiosyncratic
dollar variance in a film's payoff. Additionally, the research explores the interrelationship between
expected returns and dollar volatility, illustrating that heightened volatility corresponds to
increased anticipated returns.
(Parc, 2021) turns the spotlight on the effects of business integration within the Korean film
industry across different epochs. The study underscores the positive impact of a business-
conducive environment during integration on the industry's competitive landscape. The study's
findings hold significance for policymakers striving to shape effective cultural policies within the
film industry.
(Zhang & Pelton, 2019) delve into the media representation of Wanda Group's investments in the
U.S. entertainment sector amid the U.S.-China "trade war" during the Trump Administration. The
research uncovers three core themes embedded in media coverage: business-related aspects,
attitudes and actions of U.S. society, and China's soft power strategy. The study posits that the
alignment of business endeavors with soft power strategy plays a pivotal role in shaping public
sentiment and future relations between China and Hollywood.
(Kokas, 2020) sheds light on the investment dynamics within the Mainland Chinese film industry.
The study illuminates the distinctive molecular structure characterizing national film investment,
driven by the interplay between factors attracting commercial capital investment and those driving
state centralization. This intricate investment pattern features diverse hubs interconnected by
complementary bonds, shaped by regulatory frameworks, institutions, built environments, and
access to capital.
Collectively, these works contribute significantly to the understanding of various aspects of the
film industry, such as investment behavior, market conditions, and the influence of policy and
reputation on its competitiveness. Researchers and policymakers can benefit from the insights
provided by these studies to make informed decisions and foster growth in the film sector. Table
1 provides a summary of Related works.
Table 1: summary of Related works
Year
Topic
Methodology
Key Results
2023
Data-Driven
Decision-Making
in the Korean Film
Industry
XAI Model (DRECE
Framework) with
Dimension Reduction,
Clustering, and
Classification
DRECE Framework integrates XAI for transparent
decision-making.
Supports film professionals in enhancing box office
performance.
Empowers strategic navigation of the Korean film market.
2020
Collaboration
Networks in Film
Production
Network Analysis of
Hungarian Feature Films
(1990-2009)
Central creators acting as intermediaries enhance
creative success
Bridging the gap between core and peripheral
collaborators is beneficial
Central creators who connect core and periphery achieve
award-winning outcomes.
2022
Predicting
Derivatives Market
for Films
"WE Model" with
Machine Learning
Algorithms
WE Model predicts movie merchandising markets
accurately (72.5% accuracy).
Machine learning helps forecast movie derivatives
markets based on film attributes.
Data-driven forecasting and management of
merchandising markets are feasible.
2020
Academic-
Industry
Collaboration for
Film and TV
Production
Case Study Analysis of
Academic-Industry
Partnerships
Collaborative models enhance practical relevance of
academic programs
Identify and nurture new talent in film and TV industry
Alignment between engagement and institutional
objectives is crucial.
2012
Influence of
Reputation
Dimensions on
Distributor
Investments
Empirical Research on
Reputation and
Investment Behavior
Favorable reputation in congruent dimension positively
impacts investor behavior
Favorable reputations in incongruent dimensions weaken
this effect
Reputational category spanning" phenomenon explained.
2018
Protectionist
Policies and
Korean Film
Industry
Analysis of Protectionist
Measures and Industry
Prosperity
Protectionist policies like import quotas and subsidies
have limited impact
Market-friendly conditions and business activities are
more influential
Industry competitiveness is enhanced by business-
friendly conditions.
2015
Returns on
Tangible
Investments in the
Film Industry
Research on Expected
Returns and Dollar
Variance
Positive correlation between expected returns and film's
payoff variance
Increased volatility corresponds to heightened
anticipated returns
Expected returns linked to dollar volatility in film industry.
2021
Business
Integration Effects
in the Korean Film
Industry
Study on Business
Integration and
Competitive Landscape
Positive impact of business-conducive environment
during integration
Significant for shaping effective cultural policies in film
industry
Business-friendly conditions enhance industry
competitiveness.
2019
Media
Representation of
Wanda Group's
Investments
Analysis of Media
Coverage Amid U.S.-
China Trade War
Three core themes in media coverage: business aspects,
U.S. attitudes, China's soft power
Business alignment with soft power strategy shapes
public sentiment
Media coverage influences China-Hollywood relations
amid trade tensions.
2020
Investment
Dynamics in
Mainland Chinese
Film Industry
Study on Investment
Structure and
Centralization
National film investment driven by factors attracting
commercial capital
Interplay of factors driving state centralization shapes
investment
Distinctive molecular structure of film investment driven
by various factors.
Our article builds on the existing body of research by analyzing and examining the cooperation
and joint investments of different countries in the film industry through the technique of analyzing
social networks and themes in films. This innovative approach differs from previous similar works
as it focuses on the interconnectedness and collaborations between countries in the film sector,
shedding light on how these relationships impact the industry's competitiveness and growth. By
exploring social networks and themes, the article offers a unique perspective on the global
dynamics of the film industry, providing valuable insights for both researchers and policymakers
aiming to enhance international cooperation and investment in this sector.
3. Methodology
The steps of this research are based on Figure 3 and are as follows:
1. Data Collection:
The first step of our study involves collecting data related to movies, including their year
of release, country of production, keywords, and other relevant information from the IMDB
website. To efficiently gather this data, we employ the web crawling technique using
Python programming language. The web crawler is programmed to extract movie data
based on specified criteria such as the movie's production year and country of origin. The
collected data will serve as the foundation for our subsequent analyses.
2. Matrix Creation:
Once the data collection process is completed, we proceed to create a communication
matrix that represents the relationships between different countries in the film industry.
This matrix is constructed using common keywords found in the films produced by various
countries. The rows and columns of the matrix represent the names of the countries, while
the matrix cells contain values indicating the number of shared keywords between the
productions of the respective countries.
The fundamental principle underlying the matrix creation is that whenever two countries
share a common keyword in their movie productions, we assign a value of one in the
corresponding cell of the matrix. As a result, the more shared keywords two countries have,
the higher their allocated number in the matrix will be. This communication matrix will
provide valuable insights into the content affinity and potential cooperation between
different countries in the film industry.
3. Network Formation:
Building upon the communication matrix from the previous step, we proceed to construct
a communication network that visually represents the content affinity and cooperation
potential between countries in the film industry. For the network visualization, we utilize
the powerful Gephi software, widely recognized for its capacity in social network analysis.
Gephi allows us to explore, analyze, and present the network in a visually appealing
manner, making it an optimal choice for our research.
Using the communication matrix as input, Gephi generates a network graph, where
countries are represented as nodes, and the connections (edges) between the nodes reflect
the degree of shared keywords and content affinity between the respective countries. The
network analysis will enable us to identify clusters or communities of countries that
frequently collaborate or share similar themes in their film productions.
4. Network Analysis:
At this stage, we delve into analyzing the network at both micro and macro levels. At the
micro level, we focus on the properties of individual nodes (countries) in the network, such
as their degree centrality, betweenness centrality, and closeness centrality. These measures
help us understand the importance of each country in terms of its connectivity, influence,
and ability to act as a bridge between other countries in the network.
On the macro level, we examine global network properties, such as overall network
density, Degree, and Edges. These metrics give us insights into the overall structure of the
network and the presence of tightly knit communities or clusters within it.
As part of the network analysis, we employ community detection algorithms to identify
cohesive groups of countries within the communication network. Community detection
helps us identify clusters of countries that frequently collaborate or share thematic
similarities in their film productions. Understanding these communities will shed light on
the existing patterns of cooperation and joint investments among countries in the film
industry.
5. Policy Suggestions:
Based on the findings from the previous steps, we draw meaningful conclusions about the
cooperation and joint investments of different countries in the film industry. We present a
set of policy suggestions and recommendations that can foster and strengthen international
collaborations in filmmaking. These suggestions will be grounded in the empirical
evidence obtained from our social network analysis and community detection, aiming to
improve cross-border partnerships, boost the exchange of ideas, and encourage mutual
investments to enhance the global film landscape.
By meticulously following these steps, our research provides a comprehensive and data-driven
analysis of international cooperation in the film industry, utilizing techniques from social network
analysis to unravel the intricate relationships and thematic patterns that shape the world of cinema.
Figure 3: Research Methodology
Movie info
Collection
Data Collection Matrix Creation Network Formation Network Analysis Suggestion
Countries Network
• Community Analysis
• Analysis of Network Macro
Measures (Degree , Density)
• Analysis of Network Micro
Measures (Degree Centrality
, Betweenness Centrality ,
Closeness Centrality, Etc. )
4. Results
4.1. Data Collection
In this study, movie data from the IMDB website was collected using Python web crawling
libraries. Due to the extensive number of movies available on the platform, a filtering process was
implemented to include only those movies with a score of more than 1000. As a result, a substantial
dataset of 25,987 movies was successfully crawled. For a visual representation of the distribution
of these crawled movies across different years, please refer to Figure 4.
Figure 4: Distribution of crawled movies by year of production
During the keyword collection phase, a vast dataset comprising 3 million keywords was amassed.
To enhance the precision and relevance of our analysis, a rigorous criterion was employed,
focusing on keywords with a minimum of 3 user likes and fewer dislikes than likes. As a result of
this stringent selection process, a total of 2.1 million unique keywords were retained for the
subsequent stages of the study.
To further improve the relationship between keywords and facilitate a more comprehensive
analysis, lemmatization was performed on this subset of unique keywords. Lemmatization is a text
mining technique that involves transforming words to their base or root form, enabling us to group
related words together and reduce the dimensionality of the dataset(Shumaly et al., 2021). This
step resulted in a refined set of 38,000 keywords, which were thoroughly examined to explore
meaningful relationships and patterns in the data.
4.2. Matrix Creation
In this phase of the research, we utilized the extracted keywords from movies to construct a matrix
representing the relations between countries. The process involved cleaning and performing
lemmatization on the extracted keywords. Subsequently, if a common keyword was found in two
distinct movies, a connection (score) was established between the corresponding countries. The
connection score between the countries was then incremented with each additional occurrence of
this shared keyword in different movies.
For instance, if the keyword "violence" appeared in two films, one produced by the United States
and the other by the United Kingdom, a connection was established between the United States and
the United Kingdom, and a value of one was entered in the communication matrix. If this keyword
recurred in 17 films from each country, the final number "17" was added to the matrix to indicate
the strength of the connection between the two countries based on this keyword. This process was
repeated for all relevant keywords, resulting in the creation of a comprehensive table similar to
Figure 5, illustrating the interconnectedness between countries based on the presence of common
keywords in their respective movies.
Figure 5: An example of the created matrix
4.3. Network Formation
In the previous phase, a matrix was generated to establish and visualize the communication
network among countries, with a specific focus on shared concepts in film productions. The
outcomes of this phase are portrayed in Figure 6. The network incorporates 150 countries,
interconnected through a total of 7,800 links between country nodes, reflecting an average degree
of 52. This emphasizes a notable degree of collaboration and convergence in thematic elements
across cinematic works. The network demonstrates consistency and unity, as indicated by a density
of 0.35, thus highlighting a substantial level of connectivity within the global film industry.
Furthermore, the presence of a single connected component within the network, encompassing all
150 countries, signifies a cohesive and integrated network structure.
Figure 6:Conceptual network of countries
4.4. Network Analysis:
4.4.2. Network Micro Measures:
Applying centrality measures provides a powerful technique for micro-level examination of
networks. These metrics identify the most significant and influential components in a network
based on different criteria. As shown in Table 2, the top 20 countries ranked by degree,
betweenness, closeness, and eigenvalue centrality reveal the nations that hold critical positions and
substantial influence within the network. Such countries serve pivotal roles and strongly shape the
dynamics of the overall network. Focusing further analysis on these high centrality countries will
offer crucial insights into the underlying mechanisms and relationships governing the network.
The centrality metrics spotlight the key nodes that deserve greater attention when seeking to grasp
the nuances of the broader collaborative ecosystem.
Table 2: top 20 in centrality Measures
Degree centrality
Betweenness centrality
Closeness centrality
Eigen centrality
United States
United States
United States
United States
United Kingdom
China
China
China
France
United Kingdom
United Kingdom
United Kingdom
Germany
Japan
Japan
Japan
Japan
France
France
France
China
India
India
India
Canada
Germany
Germany
Germany
India
South Korea
South Korea
South Korea
South Korea
Canada
Canada
Canada
Italy
Australia
Australia
Australia
Spain
Italy
Italy
Italy
Australia
Spain
Spain
Spain
Brazil
Mexico
Mexico
Mexico
Mexico
Hong Kong
Hong Kong
Hong Kong
Russia
Belgium
Belgium
Belgium
Sweden
Netherlands
Netherlands
Netherlands
Poland
Poland
Norway
Poland
Belgium
Hungary
Denmark
Sweden
Denmark
Denmark
Ireland
Brazil
Netherlands
Ireland
Hungary
Russia
analyzing the centrality indices in this network significant insights into the cooperation and
investment capacities of different nations.
Degree centrality
Degree centrality measures the number of connections a node (country) has in the network.
In this case, it represents the number of common themes shared between countries' films.
The United States, with the highest degree centrality, implies it has collaborated or shared
common film themes with a significant number of countries. This could be due to its large
film industry and influence on global cinema trends.
Betweenness centrality
Betweenness centrality identifies countries that act as bridges or intermediaries between
other countries in the network. The United States, once again, holds the highest
betweenness centrality, suggesting it plays a crucial role in connecting other countries due
to its extensive collaborations. China and the United Kingdom also have high betweenness
centrality, indicating their influence in facilitating connections.
Closeness centrality
Closeness centrality measures how quickly a country can reach other countries in the
network. The United States, with the highest closeness centrality, indicates that it has
relatively short paths to reach other countries through shared film themes. China and the
United Kingdom also have high closeness centrality, reflecting their efficient access to
other countries.
Eigenvector centrality
Eigenvector centrality considers not only a country's connections but also the quality of its
connections (i.e., how connected its connections are). The United States has the highest
eigenvector centrality, indicating that it is connected to other influential countries. China,
the United Kingdom, and Japan also score high in eigenvector centrality, suggesting they
have connections to countries that themselves have many connections.
The countries with the highest centrality indices, including the United States, China, the United
Kingdom, France, and Japan, showcase their prominence and influence in the global film
production network. These countries are not only prolific film producers but also actively
collaborate and share common themes with a diverse range of other countries.
The United States consistently ranks at the top in all centrality measures, reinforcing its status as
a global film production hub. It has numerous connections, acts as a bridge between countries, is
efficiently reachable, and has quality connections.
China's high centrality indices indicate its increasing importance in the global film industry. Its
influence in bridging gaps between other countries is noteworthy. The United Kingdom's centrality
emphasizes its historical ties to global cinema and its continuing role in connecting different film
markets. France and Japan's strong centrality scores underline their cultural significance and their
ability to engage with a broad spectrum of nations.
These high centrality index countries not only contribute significantly to the film production
landscape but also serve as key players in fostering international collaboration and investment in
the industry. Their centrality underscores their potential to shape trends, share expertise, and drive
innovation in global cinema.
4.4.3 Community Detection and Analysis:
Community detection is a powerful analytical tool that allows researchers to identify coherent
groups in a network based on the connections between its nodes. By applying community detection
algorithms to the network, we can discover distinct clusters of countries with similar creative
endeavors and thematic approaches in their filmmaking. By identifying these clusters, we can gain
a deeper understanding of the connectivity and potential for collaboration in the global film
production industry. Figure 7 shows a representation of the result of discovering associations in
the network.
Figure 7: Communication network of countries after community detection
Table 3 shows the details of the countries present in each forum and the most frequent concepts in
the products of that forum.
Table 3: The main countries present in each community along with the most frequent keywords
color
Nodes
edges
density
The most important
countries
Keywords
39
107
0.14
England, Canada, India,
Australia, Hungary, Poland,
Ireland, Czech Republic,
South Africa, New Zealand
Confrontation, Gunfight, Near
Death Experience, Apology,
Surrealism, Lawyer, Revolver
53
280
0.20
United States, France,
Germany, Belgium, Italy,
Switzerland, Netherlands,
Austria, Russia, Turkey
Politics, police, anger, street
market, football
16
42
0.35
Qatar, United Arab
Emirates, Iran, Egypt,
Lebanon, Palestine, Jordan,
Saudi Arabia, Iraq, Kuwait
Warrior, prayer, religion, ambush,
tent, survival, ritual
5
9
0.9
Croatia, Serbia, North
Macedonia, Slovenia
Prisoner, tragic event, comedy,
musician, magic ceremony, magic
17
54
0.39
Spain, Mexico, Brazil,
Argentina, Chile, Portugal,
Colombia, Uruguay, Peru
Cemetery, Near Death Experience,
Apology, Ambush, Priest
6
13
0.86
Sweden, Norway, Denmark,
Finland, Iceland
Final Race, Apartment, Underwater
Scene, Surrealism, Jungle, Sniper,
Priest, Non-linear Story, Motor
Vehicle, Confrontation,
Interrogation
15
47
0.44
Japan, China, Hong Kong,
South Korea, Thailand,
Taiwan, Philippines,
Singapore, Malaysia
Nightmare, interrogation, airport,
priest, nurse, urban environment
Based on the countries and keywords in each community, we can name and analyze each
community as follows:
Community 0: Global Thrill Seekers
This community consists of countries such as England, Canada, India, Australia, and more.
These countries are connected by their shared focus on intense and action-packed film
themes like confrontation, gunfights, and near-death experiences. The choice of "Global
Thrill Seekers" as the name represents the excitement and adrenaline rush these countries
bring to their films.
Members of this community could collaborate on international co-productions that
emphasize high-stakes action and suspense. Shared investment in special effects, stunts,
and production equipment can lead to visually stunning films with global appeal.
Consider establishing an international fund to support co-produced action films, with
participating countries contributing resources based on their strengths. This could enable
the creation of big-budget productions that attract audiences worldwide.
Community 1: Socio-Political Cinema
This community includes the USA, France, Germany, Belgium, and more. The chosen
name reflects their shared theme of politics and cultural interactions portrayed in their
films. These countries often delve into societal issues, employing films as a medium to
address global concerns.
Members of this community can collaborate on films that promote cross-cultural
understanding and tackle global challenges. Co-productions can showcase diverse
perspectives, fostering a sense of unity in addressing complex issues.
Create an annual international film festival focused on films from this community that deal
with diplomacy, politics, and societal discourse. This festival can serve as a platform for
discussions, fostering deeper connections among the member countries.
Community 2: Faith and Unity
This community comprises countries like Qatar, United Arab Emirates, Iran, Egypt, and
more. The name reflects the prevalent religious and cultural themes in their films,
highlighting the importance of faith and unity in their societies.
Collaborative projects could focus on exploring religious and cultural diversity, fostering
understanding and tolerance. Joint investment in script development and cultural
consultants can ensure authenticity and respectful portrayal.
Establish an international fund to support films that promote interfaith dialogue and
understanding. These films can shed light on the commonalities and shared values among
diverse cultures.
Community 3: Balkan Connections
This smaller community consists of countries like Croatia, Serbia, North Macedonia, and
Slovenia. The name signifies their shared history and geographic proximity. Despite their
smaller size, these countries collaborate on projects that highlight their shared heritage.
Collaborative efforts can focus on producing historical dramas, showcasing their
intertwined histories and cultural identities. Joint funding for research and script
development can lead to authentic and compelling stories.
Organize a regional film fund specifically for the Balkan countries, enabling them to jointly
finance and create films that celebrate their shared cultural heritage while appealing to
international audiences.
Community 4: Latin Elegy
This community consists of countries like Spain, Mexico, Brazil, Argentina, and more. The
chosen name reflects the shared themes of passion and emotion prevalent in their films,
often depicted through elements like cemeteries, near-death experiences, and apologies.
Members can collaborate on films that capture the rich tapestry of emotions and cultural
nuances within their respective societies. Co-productions can blend different storytelling
styles for unique and impactful narratives.
Create a fund dedicated to producing films that celebrate Latin American culture and
emotions. These films can emphasize emotional storytelling and artistic expression while
promoting cross-cultural awareness.
Community 5: Nordic Odyssey
This community includes countries like Sweden, Norway, Denmark, Finland, and Iceland.
The chosen name reflects the shared themes of natural landscapes and surrealism present
in their films, often depicted through scenes like final races, underwater sequences, and
non-linear storytelling.
Collaborations can focus on producing visually captivating films that highlight the
stunning Nordic landscapes and unique narrative structures. Shared investments in
cinematography and post-production can elevate the visual appeal of their films.
Establish a production fund that supports films set in Nordic environments, promoting the
region's scenic beauty and distinctive storytelling. This fund could encourage joint efforts
in scouting locations and utilizing local talent.
Community 6: Asian Enigma
This community comprises countries like Japan, China, Hong Kong, South Korea, and
more. The name reflects the intriguing and mysterious film themes prevalent in their works,
often featuring nightmares, interrogations, and urban environments.
Collaborative projects can explore themes that delve into the complexity of modern Asian
societies and their urban environments. Shared resources in technology and expertise can
enhance the visual and narrative impact of their films.
Create an Asian film co-production initiative that encourages filmmakers from this
community to collaborate on projects that showcase the diversity and dynamism of Asian
cultures while captivating global audiences.
5. Conclusion
This study conducted an analysis of the communication network among 150 countries based on
common themes found in their film productions. The findings of this research revealed a highly
interconnected global network with substantial cooperation and convergence in cinematic
concepts.
The network analysis identified countries with the highest influence, as demonstrated by their high
centrality scores. The United States, China, and England emerged as key players, reflecting their
prominence in driving and pioneering innovative trends in the industry. Their centrality
underscores their leadership in fostering international collaboration and investment.
The identification of communities unveiled distinct clusters of countries united by common
creative approaches. For instance, the "Global Thrill Seekers" community produces action films,
while the "Socio-Political Cinema Society" delves into social issues. Each community offers
opportunities for targeted collaboration and investment based on shared interests.
Broadly, this research highlights contributions and growth potential in the global film production
landscape. As national cinemas increasingly converge through collaborative investments, joint
productions, and shared creative themes, new prospects for strategic unity and investment
initiatives emerge. Regionally appropriate budgets, collaborative production agreements, film
festivals, and network platforms can all contribute to enhancing these connections.
Utilizing network analysis and community identification techniques, stakeholders can make
informed decisions regarding focal points for collaboration and investment to maximize impact.
Just as the network evolves over time, further analysis can identify growing capacities and
emerging creative communities.
Policy Recommendations:
• Drawing insights from social network analysis and the article's findings, here are some
policy recommendations to enhance collaborative efforts and investments in the global film
industry:
• Establish international joint production funds to financially support film projects that align
with creative or thematic interests shared between countries. Such funds can facilitate the
creation of high-quality, globally appealing productions through shared resources.
Targeted budgets could be allocated for identified communities like "Global Thrill
Seekers" or "Faith and Unity."
• Initiate cultural exchange programs to promote the exchange of talents, ideas, and best
practices among countries through workshops, educational programs, residencies, etc. This
approach can enrich skills and perspectives.
• Launch region-specific film funds dedicated to projects supporting cultural heritage and
narratives in specific regions like the Middle East, Northern Europe, the Balkans, Latin
America, etc. This approach encourages local collaboration.
• Organize international film festivals with a focus on outstanding productions from
identified communities. These events provide platforms for showcasing films, networking,
and discussions.
• Develop joint marketing and distribution strategies to promote films from a community
globally. This enhances accessibility and influence through coordinated efforts.
• Standardize policies for collaborative production, incentive schemes, and treaty
frameworks to enable seamless cross-border cooperation between countries.
• Establish joint training programs between national film institutions to develop specialized
skills necessary for international productions.
• Create databases and platforms for sharing information about best practices in collaborative
production, success stories, upcoming projects, etc., to enhance accessibility.
• Generate incentives for private investment funds specifically geared towards financing
international joint productions, especially from key identified countries. This incentivizes
new capital resources.
In conclusion, this study's comprehensive analysis sheds light on the potential of global film
production collaborations and growth. As the boundaries between national cinemas blur, fostering
strategic alliances and investment becomes paramount. The proposed policy recommendations aim
to guide stakeholders in harnessing the power of collaborative creativity to shape the future of the
global film industry.
References
Bedi, P., & Sharma, C. (2016). Community detection in social networks. Wiley Interdisciplinary Reviews:
Data Mining and Knowledge Discovery, 6(3), 115-135.
Behrens, R., Foutz, N. Z., Franklin, M., Funk, J., Gutierrez-Navratil, F., Hofmann, J., & Leibfried, U. (2021).
Leveraging analytics to produce compelling and profitable film content. Journal of Cultural
Economics, 45, 171-211.
Betzler, D., & Leuschen, L. (2021). Digitised value chains in the creative industries: Is there a
convergence of Swiss film and game production? Creative Industries Journal, 14(3), 226-244.
Bulgurcu, B., Van Osch, W., & Kane, G. C. (2018). The rise of the promoters: user classes and
contribution patterns in enterprise social media. Journal of management information systems,
35(2), 610-646.
Dašić, D., & Kostadinović, G. (2022). Film Industry in the Function of Destination Branding. 6th
International Thematic Monograph Modern Management Tools and Economy of Tourism Sector
in Present Era. UDEKOM, 441-456.
Dastidar, S. G., & Elliott, C. (2020). The Indian film industry in a changing international market. Journal of
Cultural Economics, 44, 97-116.
Ding, Y., Wu, C., Liu, X., Zou, Y., & Zhou, P. (2022). WE model: A Machine Learning Model Based on Data-
Driven Movie Derivatives Market Prediction. arXiv preprint arXiv:2207.02423.
Ding, Z., Liu, R., & Yuan, H. (2021). A text mining-based thematic model for analyzing construction and
demolition waste management studies. Environmental Science and Pollution Research, 28(24),
30499-30527.
Ebbers, J. J., & Wijnberg, N. M. (2012). The effects of having more than one good reputation on
distributor investments in the film industry. Journal of Cultural Economics, 36(3), 227-248.
https://doi.org/10.1007/s10824-012-9160-z
El-Moussaoui, M., Agouti, T., Tikniouine, A., & Adnani, M. E. L. (2019). A comprehensive literature review
on community detection: Approaches and applications. Procedia Computer Science, 151, 295-
302. https://doi.org/10.1016/j.procs.2019.04.042
Guo, P. (2023). Research the Development of International Joint Film Production in the Context of the
New Era. Studies in Art and Architecture, 2(2), 16-20.
HabibAgahi, M. R., Kermani, M. A. M. A., & Maghsoudi, M. (2022). On the Co-authorship network
analysis in the Process Mining research Community: A social network analysis perspective.
Expert Systems with Applications, 206, 117853.
Jalilvand Khosravi, M., Maghsoudi, M., & Salavatian, S. (2022). Identifying and Clustering Users of VOD
Platforms Using SNA Technique: A case study of Cinemamarket. New Marketing Research
Journal, 11(4), 20-21.
Johnsen, M. (2023). The Art of Investing in Film: Strategies for Success. Maria Johnsen.
Juhász, S., Tóth, G., & Lengyel, B. (2020). Brokering the core and the periphery: Creative success and
collaboration networks in the film industry. PloS one, 15(2), e0229436.
Kermani, M. A., Maghsoudi, M., Hamedani, M. S., & Bozorgipour, A. (2022). Analyzing the
interorganizational collaborations in crisis management in coping with COVID-19 using social
network analysis: Case of Iran. Journal of emergency management (Weston, Mass.), 20(3), 249-
266.
Kermani, M. A. M. A., Maghsoudi, M., Hamedani, M. S., & Bozorgipour, A. (2022). Analyzing the
interorganizational collaborations in crisis management in coping with COVID-19 using social
network analysis: Case of Iran. Journal of emergency management (Weston, Mass.), 20(3), 249-
266.
Kokas, A. (2020). Serious chemistry on set: the molecular structure of film investment in China.
International Journal of Cultural Policy, 26(4), 446-458.
https://doi.org/10.1080/10286632.2019.1614568
Kübler, R., Seifert, R., & Kandziora, M. (2021). Content valuation strategies for digital subscription
platforms. Journal of Cultural Economics, 45, 295-326.
Lee, Y. W., Moon, H.-C., & Yin, W. (2020). Innovation process in the business ecosystem: the four
cooperations practices in the media platform. Business Process Management Journal, 26(4),
943-971.
Leem, S., Oh, J., So, D., & Moon, J. (2023). Towards Data-Driven Decision-Making in the Korean Film
Industry: An XAI Model for Box Office Analysis Using Dimension Reduction, Clustering, and
Classification. Entropy, 25(4), 571. https://www.mdpi.com/1099-4300/25/4/571
Maghsoudi, M., Jalilvand Khosravi, M., & Salavatian, S. (2023). Analyzing the Ecosystem of Iranian
websites for Watching and Downloading videos based on User Behavior. Iranian Journal of
Information Processing and Management, 38(3), 1067-1094.
Maghsoudi, M., & Nezafati, N. (2023). Navigating the Acceptance of Implementing Business Intelligence
in Organizations: A System Dynamics Approach. Telematics and Informatics Reports, 100070.
https://doi.org/https://doi.org/10.1016/j.teler.2023.100070
Maghsoudi, M., Shokouhyar, S., Ataei, A., Ahmadi, S., & Shokoohyar, S. (2023). Co-authorship network
analysis of AI applications in sustainable supply chains: Key players and themes. Journal of
cleaner production, 138472. https://doi.org/https://doi.org/10.1016/j.jclepro.2023.138472
Maghsoudi, M., Shokouhyar, S., Khanizadeh, S., & Shokoohyar, S. (2023). Towards a taxonomy of waste
management research: An application of community detection in keyword network. Journal of
cleaner production, 136587. https://doi.org/https://doi.org/10.1016/j.jclepro.2023.136587
Maghsoudi, M., & Shumaly, S. Study of co-authorship network of papers in all sessions of IIIEC
Conference.
Mateer, J. (2020). Academic-Industry Collaboration for Commercial Film and Television Production: an
exploration of case studies. 2019 International Conference on Global Cultural and Creative
Industries: Industry-Academy Cooperation and the Opportunities in China,
Messerlin, P., & Parc, J. (2018). In search of an effective trade policy for the film industry: Lessons from
Korea. Journal of World Trade, 52(5).
Newman, M. E. (2006). Modularity and community structure in networks. Proceedings of the national
academy of sciences, 103(23), 8577-8582.
Parc, J. (2018). Evaluating the Effects of Protectionism on the Film Industry: A Case Study Analysis of
Korea. In P. C. Murschetz, R. Teichmann, & M. Karmasin (Eds.), Handbook of State Aid for Film:
Finance, Industries and Regulation (pp. 349-366). Springer International Publishing.
https://doi.org/10.1007/978-3-319-71716-6_20
Parc, J. (2021). Business integration and its impact on film industry: The case of Korean film policies from
the 1960s until the present. Business History, 63(5), 850-867.
Powers, T. (2015). Expected returns on real investments: Evidence from the film industry. Available at
SSRN 2460108.
Said, A., Abbasi, R. A., Maqbool, O., Daud, A., & Aljohani, N. R. (2018). CC-GA: A clustering coefficient
based genetic algorithm for detecting communities in social networks. Applied Soft Computing,
63, 59-70. https://doi.org/https://doi.org/10.1016/j.asoc.2017.11.014
Shumaly, S., Yazdinejad, M., & Guo, Y. (2021). Persian sentiment analysis of an online store independent
of pre-processing using convolutional neural network with fastText embeddings. PeerJ Computer
Science, 7, e422.
Vlassis, A. (2016). Soft power, global governance of cultural industries and rising powers: the case of
China. International Journal of Cultural Policy, 22(4), 481-496.
Vogel, H. L. (2020). Entertainment industry economics: A guide for financial analysis. Cambridge
University Press.
Yap, M., Luo, D., Cats, O., van Oort, N., & Hoogendoorn, S. (2019). Where shall we sync? Clustering
passenger flows to identify urban public transport hubs and their key synchronization priorities.
Transportation Research Part C: Emerging Technologies, 98, 433-448.
Yayla, S., Kutlubay, O. C., Cicek, M., & Yeniyurt, S. (2023). Once upon a time in a foreign market: The role
of cultural distance in the economic performance of multilateral non-equity partnerships.
International Business Review, 32(4), 102139.
Zhang, X., & Pelton, L. (2019). Exploring the themes in the US media coverage of Wanda’s investments in
Hollywood. Journal of Media Business Studies, 16(1), 58-76.
Zohdi, M., Maghsoudi, M., & Nooralizadeh, H. (2022). Providing a User-Based Behavior Model to
Recommend a Movie Using the Social Network Analysis (Case Study: CinemaMarket). Sciences
and Techniques of Information Management, 8(1), 451-484.
