Abstract
<jats:p>Analysis of Recent Research and Publications. Recent studies on malicious or inauthen-tic activity in social networks have progressed from account-level bot detection toward group-level identification of coordinated behavior. While individual classification remains important, it often fails to capture organized campaigns where coordination across many ac-counts is the primary operational signal. This shift has encouraged community-oriented ap-proaches that combine interaction graphs with temporal synchronization signals and shared interaction elements, and that apply statistical testing to distinguish coordination from back-ground synchronicity caused by daily rhythms and event-driven bursts. Purpose of the Study. The purpose of this study is to develop and validate a community-level, software-oriented methodology for detecting and quantifying coordinated activity among bot-programs. The methodology integrates interaction-network structure, temporal regularities of activity, and shared interaction elements, and produces outputs that are suit-able for automated monitoring pipelines. The study aims to generate interpretable community rankings and to extract statistically significant coordinated pairs that can serve as explana-tory artifacts for analysts and downstream risk-monitoring modules. Main Content of the Study. The study constructs a weighted interaction network in which nodes represent accounts and edges aggregate multi-type interaction intensity between account pairs. Communities are detected via modularity-based optimization and refined with programmatic filtering rules to ensure internal connectivity and stable results across re-peated runs. Within each community, coordination is quantified by combining two signals: temporal synchronization derived from binned activity sequences with limited time shifts, and similarity over shared interaction elements such as repeated posts, mentions, hashtags, and link domains. To separate genuine coordination from incidental synchronicity, statistical sig-nificance is evaluated using a permutation-based null model that preserves each account’s activity volume, while multiple comparisons are controlled to limit false discoveries. Each community is then summarized by a robust measure of coordination strength among signifi-cant pairs and by the proportion of significant pairs within the community, which enables ranking and prioritization. The implemented pipeline also returns compact subgraphs of sig-nificant pairs, allowing traceable and reproducible explanations of why a community is ranked highly. Conclusions. The empirical evaluation indicates that the proposed coordination meas-ures separate highly coordinated bot-program communities from background groups, and that permutation-based testing reduces false positives caused by incidental synchronicity. Ranking by coordination concentrates bot-program prevalence in the top segment of commu-nities and yields compact, interpretable subgraphs of statistically significant coordinated pairs that support analyst-oriented investigation and automated monitoring. Overall, the study demonstrates a reproducible software pipeline that bridges community detection and statistically validated coordination scoring, providing actionable outputs for monitoring co-ordinated information campaigns.</jats:p>