=Paper=
{{Paper
|id=Vol-1794/afcai16-paper3
|storemode=property
|title=Affective Character Network for Understanding Plots of Narrative Contents
|pdfUrl=https://ceur-ws.org/Vol-1794/afcai16-paper3.pdf
|volume=Vol-1794
|authors=O-Joun Lee,Jason Jung
|dblpUrl=https://dblp.org/rec/conf/afcai/LeeJ16
}}
==Affective Character Network for Understanding Plots of Narrative Contents==
https://ceur-ws.org/Vol-1794/afcai16-paper3.pdf
Affective Character Network for Understanding
Plots of Narrative Multimedia Contents
O-Joun Lee1 and Jason J. Jung1?
Department of Computer Engineering
Chung-Ang University
Seoul, Korea 156-756
{concerto34,j3ung}@cau.ac.kr
Abstract. It is important to understand the stories from narrative mul-
timedia contents (e.g., movies), and to exploit the stories for smart ser-
vices (e.g., video summarization and personalized multimedia recommen-
dation). In this paper, we extend CharNet (Character Network) to Af-
fective CharNet (Affective Character Network) by annotating emotional
relationships between characters. More importantly, we propose a novel
method on analyzing Affective CharNet for extracting the plots from the
narrative multimedia contents.
Keywords: Character Network, Affective Computing, Narrative Anal-
ysis, Social Network Analysis, Emotional Social Network
1 Character Network
The goal of this study is to understand the stories (more properly, plots) of
narrative multimedia contents by using Affective CharNet (Affective Character
Network). To model and analyze stories and plots of narrative multimedia con-
tents, various studies have been conducted with social network analysis (e.g.,
RoleNet [7], CharNet [4], and CoCharNet [6, 5]). Common contribution of these
heuristics-based studies are based on social networks between characters. They
have measured the strength of social ties between characters based on frequencies
of co-occurrences and dialogues of characters. Based on the intensities of social
relationships, they classified characters into main characters, minor characters,
and extras.
Definition 1 (Character Network). A character network N can be repre-
sented as
N = hCH, Ri (1)
where C is a set of characters in the story, and R ⊆ |C| × |C|.
However, they are not able to clearly reflect the plots or storylines. The plot
is defined as a sequence of events which are described in contents and logically
?
Corresponding author
2 Lee and Jung
related with each other. As an example, CoCharNet (our previous study [6]) can
partially represent external shapes of stories. The social relationships are simply
accumulative based on frequencies of co-occurrence or dialogues. Even though
social relationships are discovered, it is hard to represent events between char-
acters over time. Thereby, CoCharNet needs to be extended to analyze dynamic
changes of social relationships.
Another issue on the previous studies is that they assume that all the so-
cial relationships among characters are homogeneous. Jung et al. [1] also tried
to annotate emotional states of characters on CharNet. However, they are not
able to reflect dynamic changes of emotional relationships, since they only have
considered emotional states of characters on average.
2 Affective CharNet for Extracting Plots
By considering emotional relationships between characters, the character net-
work N can be extended to an affective character network.
Definition 2 (Affective Character Network). An affective character net-
work Af N can be represented as
Af N = hCH, E, Ri (2)
where C and R is a set of characters in the story and a set of emotions, respec-
tively. The social relationships are represented as R ⊆ |C| × |E| × |C|.
To annotate dynamic changes of emotional relationships and detect events
between characters, we apply the following two approaches.
– first, segmenting contents based on co-occurrence of characters, and
– second, detecting events based on radical changes of emotional relationships.
Based on these approaches, plots are exploited by following procedures.
– segmenting contents and annotating emotional relationships between char-
acters at each segment,
– detecting the events described in contents by searching radical changes of
emotional relationships,
– and modeling plots based on the events and characters involved in the events.
Prior to annotating emotional relationships on CharNet, we have to extract
emotional states of characters at each segment. Data sources to extract emotional
states are different form kinds of contents. If they are visual contents, we can
apply facial expressions. In other case, if they are textual contents, we can use
emotional words. Preliminarily, we have manually annotated emotional states as
a 1-dimensional value from −1 to 1. If the value is close to −1, it means negative
emotional states. In opposite case, it means positive ones.
To transform emotional states into emotional relationships, we have made
three assumptions that emotion of a character will be directed to characters
Affective Character Network for Understanding Plots 3
– which have high connectivity with him/her,
– which are included in same social groups with him/her, and
– which have connection with him/her.
Fig 1 is an example of the Affective CharNet, where cha indicates an a-th
event and Ea,b means emotional relationship of cha toward chb .
Fig. 1. An example of affective CharNet
To model and visualize plots of the narrative multimedia contents, we locate
detected events in order which is described in contents. Also, by annotating
characters involved in each event, we make the model enable to represent logical
linkages between characters. Fig. 2 is an example of the proposed computational
model of the plots, where ei indicates an i-th event.
3 Conclusion
In this study, we propose a method to annotate emotional relationships on Char-
Net. Also, we extract and model plots of narrative multimedia contents. It is
meaningful in terms of the first try for computational analysis of plots.
However, this is a manifesto paper with some preliminary research with many
practical limitations. We have planned the following work in future.
– Most importantly, we need to conduct experimental study to prove the pro-
posed methods.
– More refined method to extract emotional relationships is needed.
– An appropriate way to evaluate accuracy and efficiency of the proposed
method is necessary.
4 Lee and Jung
Fig. 2. An example of the computational model of plot
– Storification [2] process will be considered to personal recommendation ser-
vices.
– Transmedia ecosystem [3] will be considered to show the relationships among
the narrative contents.
Acknowledgments. This work was supported by the National Research Foun-
dation of Korea (NRF) grant funded by the Korea government (MSIP) (NRF-
2014R1A2A2A05007154). Also, this work was supported by the Ministry of Edu-
cation of the Republic of Korea and the National Research Foundation of Korea
(NRF-2015S1A5B6037297).
References
1. Jung, J.J., You, E., Park, S.: Emotion-based character clustering for managing story-
based contents: a cinemetric analysis. Multimedia Tools Appl. 65(1), 29–45 (2013),
http://dx.doi.org/10.1007/s11042-012-1133-x
2. Jung, J.E., Hong, M., Nguyen, H.L.: Serendipity-based Storification: From Lifelog-
ging to Storytelling. Multimedia Tools and Applications. http://dx.doi.org/10.
1007/s11042-016-3682-x
3. Jung, J.E., Lee, O-J., You, E.-S., Nam, M.-H.: A Computational Model of Trans-
media Ecosystem for Story-based Contents. Multimedia Tools and Applications.
http://dx.doi.org/10.1007/s11042-016-3626-5
4. Park, S., Oh, K., Jo, G.: Social network analysis in a movie using character-
net. Multimedia Tools Appl. 59(2), 601–627 (2012), http://dx.doi.org/10.1007/
s11042-011-0725-1
5. Tran, Q.D., Hwang, D., Lee, O.-J, Jung, J.E.: Exploiting Character Networks
for Movie Summarization. Multimedia Tools Appl. http://dx.doi.org/10.1007/
s11042-016-3633-6
Affective Character Network for Understanding Plots 5
6. Tran, Q.D., Jung, J.E.: Cocharnet: Extracting social networks using character co-
occurrence in movies. Journal of Universal Computer Science 21(6), 796–815 (2015),
http://www.jucs.org/jucs_21_6/cocharnet_extracting_social_networks
7. Weng, C., Chu, W., Wu, J.: Rolenet: Movie analysis from the perspective of social
networks. IEEE Trans. Multimedia 11(2), 256–271 (2009), http://dx.doi.org/10.
1109/TMM.2008.2009684