Analisis Jaringan Interaksi Instagram pada Peristiwa Ponpes Al Khoziny Menggunakan Social Network Analysis dan Algoritma Clauset-Newman-Moore

Arianto, Farhan Dequika Pradana Putra (2026) Analisis Jaringan Interaksi Instagram pada Peristiwa Ponpes Al Khoziny Menggunakan Social Network Analysis dan Algoritma Clauset-Newman-Moore. Undergraduate thesis, UPN Veteran Jawa Timur.

[img] Text (COVER)
21083010062_Farhan Dequika_Skripsi Final_COVER.pdf
Download (2MB)
[img] Text (BAB 1)
21083010062_Farhan Dequika_Skripsi_BAB 1.pdf
Download (160kB)
[img] Text (BAB 2)
21083010062_Farhan Dequika_Skripsi_BAB 2.pdf
Restricted to Repository staff only until 15 September 2029.
Download (384kB)
[img] Text (BAB 3)
21083010062_Farhan Dequika_Skripsi_BAB 3.pdf
Restricted to Repository staff only until 15 September 2029.
Download (594kB)
[img] Text (BAB 4)
21083010062_Farhan Dequika_Skripsi_BAB 4.pdf
Restricted to Repository staff only until 15 September 2029.
Download (1MB)
[img] Text (BAB 5)
21083010062_Farhan Dequika_Skripsi_BAB 5.pdf
Restricted to Repository staff only until 15 September 2029.
Download (141kB)
[img] Text (Daftar Pustaka)
21083010062_Farhan Dequika_Skripsi_Daftar Pustaka.pdf
Download (117kB)
[img] Text (Lampiran)
21083010062_Farhan Dequika_Skripsi_Lampiran.pdf
Restricted to Repository staff only until 15 September 2029.
Download (219kB)

Abstract

Instagram interaction networks related to the Ponpes Al Khoziny incident from September 29 to October 13, 2025, were analyzed to examine account relationships, community formation, the structural positions of key actors, and the distribution of comment sentiment. The initial data were obtained through NodeXL and subsequently processed using a Python-based GUI for data selection and cleaning, edge-weight construction, graph modeling, metric calculation, community detection, key actor identification, sentiment classification, integration, visualization, and result export. The network analysis applied Social Network Analysis using a directed weighted graph as the main graph representation. Communities were detected using the Clauset-Newman-Moore algorithm, while key actors were identified based on in-degree, betweenness centrality, and PageRank. Sentiment classification was performed on comments containing valid text using the pretrained w11wo/indonesian-roberta-base-sentiment-classifier model. After data selection and cleaning, 4,618 final interactions were obtained for network analysis, forming 3,865 nodes and 4,208 unique weighted edges with a network density of 0.000282. Community detection produced 80 communities with a modularity value of 0.872337. Combining the three ranking lists, each containing the top ten accounts for its respective metric, resulted in 19 unique key actors. Sentiment analysis of 4,632 comments identified 911 positive comments or 19.67%, 2,737 negative comments or 59.09%, and 984 neutral comments or 21.24%. Negative labels were the largest category among the analyzed comments; however, this distribution represents the model's classification results for the analyzed comment data and is not used to infer the attitudes of all Instagram users or the wider public. The results show that network structure, communities, the structural positions of key actors, and comment sentiment can be examined in an integrated manner within the data and period used in the analysis

Item Type: Thesis (Undergraduate)
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorMuhaimin, AmriNIDN0023079502amri.muhaimin.stat@upnjatim.ac.id
Thesis advisorNasrudin, MuhammadNUPTK4241774675130323nasrudin.fasilkom@upnjatim.ac.id
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76.6 Computer Programming
Divisions: Faculty of Computer Science > Departemen of Data Science
Depositing User: Farhan Dequika Pradana Putra Arianto
Date Deposited: 16 Sep 2026 04:32
Last Modified: 16 Sep 2026 04:49
URI: https://repository.upnjatim.ac.id/id/eprint/60228

Actions (login required)

View Item View Item