arXiv subjects

Arthur Capozzi

Publications and source records attributed to Arthur Capozzi.

11 recordsLinked to original sources

arXiv2026

Agentic Graph Retrieval-Augmented Generation for Auditable Commercial Registry Analysis

Public commercial registries are formally open, yet their practical analysis remains difficult because relevant facts are scattered across millions of records that combine structured metadata, multilingual legal notices, temporal events, and entity aliases. This paper presents a controlled, tool-mediated agentic GraphRAG architecture for auditable natural-language analysis of such registries. The proposed pipeline transforms publications from the Swiss Official Gazette of Commerce into a Neo4j knowledge graph comprising over five million nodes and 4.7 million relationships. It combines deterministic ingestion of structured registry fields, LLM-assisted extraction of latent actors from unstructured notices, and a deterministic identity-resolution layer. An analytical agent operates on this graph through intent routing, restricted graph tools, bounded reflection, and state-machine-guided response synthesis. We evaluate the system using a multi-tier protocol covering answer quality, retrieval behavior, entity resolution, and multi-turn conversational performance. The complete architecture is compared with dense, lexical, and hybrid flat-retrieval baselines and with controlled architectural ablations. On a manually curated benchmark, graph-mediated retrieval increases factual correctness from 0.26 for the strongest flat-retrieval baseline to 0.83 for the complete system, with comparable improvements in relevance and completeness. Ablation results show that bounded reflection improves answer quality while intent routing and LLM-based graph enrichment improve reliability in difficult entity resolution tasks. An exploratory dashboard displays the graph evidence and execution traces underlying each response, allowing users to inspect how answers were produced.

cs.IR↗

arXiv2026

Conditional Publics: Shared Events and Divergent Meanings in the European Twitter Debate on the Ukraine War

How do European publics debate a geopolitical crisis on social media, and do they inhabit a shared informational reality? We analyze over 38 million geolocated tweets from 20 European countries during the first eight months of the Russian invasion of Ukraine. Using retweet community detection and stance annotation across six issues, we identify 'hawkish' and 'doveish' opinion clusters present within almost every country studied. We find that structural polarization is driven not by radicalization, but by the exit of casual users. Crucially, whether opposing sides orient to the same events depends on the issue. On pragmatist issues, both sides react to the same high-profile events, forming an agonistic public sphere. Instead, on interpretive issues, they operate as affective publics and counterpublics constructing divergent meanings. We propose conditional publics to describe formations whose relational structure, sharing or fracturing a referential frame, depends on the epistemic character of the debated issue.

cs.CY↗

arXiv2026

Beyond the Townhall: Spatial Anchoring and LLM Agents for Scalable Participatory Urban Planning

Participatory urban planning is central to sustainable city-making, yet the technically demanding nature of such interventions often limits meaningful involvement by diverse publics. We introduce a scalable digital participation platform that embeds sustainability projects within a navigable digital twin. Citizens experience a guided virtual walkthrough with audio narration employing the method of loci and spatial anchoring to support mnemonic encoding and recall. This immersive interface is augmented by two purpose-built LLM assistants: one delivers source-grounded factual clarifications, while the other facilitates reflective discussion. We evaluated this system in a randomized controlled online experiment (N = 195) against conventional industry practices (static visualizations and text-based consultations). Results show that spatially anchored immersive presentation significantly improved information recall, which substantially shifted participants' attention from individual inconveniences to collective, community-oriented sustainability benefits. Consequently, participants provided significantly more constructive, solution-focused feedback to the (simulated) municipality. These findings establish a practical tool for cities and policymakers to foster inclusive, democratic participation in sustainability transitions.

cs.HC↗

arXiv2025

The direct democracy paradox: Microtargeting and issue ownership in Swiss online political ads

Political advertising on social media has fundamentally reshaped democratic deliberation, playing a central role in electoral campaigns and propaganda. However, its systemic impact remains largely theoretical or unexplored, raising critical concerns about institutional fairness and algorithmic transparency. This paper provides the first data-driven analysis of the relationship between direct democracy and political advertising on social media, leveraging a novel dataset of 40,000 political ads published on Meta in Switzerland between 2021 and 2025. Switzerland's system of direct democracy, characterized by frequent referenda, provides an ideal context for examining this relationship beyond standard electoral cycles. The results reveal the sheer scale of digital campaigning, with 560 million impressions targeting 5.6 million voters, and suggest that greater exposure to "pro-Yes" advertising significantly correlates with referendum approval outcomes. Demographic microtargeting analysis suggests partisan strategies: Centrist and right-wing parties predominantly target older men, whereas left-wing parties focus on young women. Regarding textual content, a clear pattern of "talking past each other" is identified; in line with the issue ownership theory, parties avoid debating shared issues, preferring to promote exclusively owned topics. Furthermore, the parties' strategies are so distinctive that a machine learning model trained only on audience and topic features can accurately predict the author of an advertisement. This article highlights how demographic microtargeting, issue divergence, and tailored messages could undermine democratic deliberation, exposing a paradox: Referenda are designed to be the ultimate expression of the popular will, yet they are highly susceptible to invisible algorithmic persuasion.

cs.SI↗

arXiv2025

Breaking the Code: Multi-level Learning in the Eurovision Song Contest

Organizations learn from the market, political, and societal responses to their actions. While in some cases both the actions and responses take place in an open manner, in many others, some aspects may be hidden from external observers. The Eurovision Song Contest offers an interesting example to study organizational level learning at two levels: organizers and participants. We find evidence for changes in the rules of the Contest in response to undesired outcomes such as runaway winners. We also find strong evidence of participant learning in the characteristics of competing songs over the 70-years of the Contest. English has been adopted as the lingua franca of the competing songs and pop has become the standard genre. Number of words of lyrics has also grown in response to this collective learning. Remarkably, we find evidence that four participating countries have chosen to ignore the "lesson" that English lyrics increase winning probability. This choice is consistent with utility functions that award greater value to featuring national language than to winning the Contest. Indeed, we find evidence that some countries -- but not Germany -- appear to be less susceptible to "peer" pressure. These observations appear to be valid beyond Eurovision.

physics.soc-ph↗

arXiv2025

FlowSeries: Anomaly Detection in Financial Transaction Flows

In recent years, the digitization and automation of anti-financial crime (AFC) investigative processes have faced significant challenges, particularly the need for interpretability of AI model results and the lack of labeled data for training. Network analysis has emerged as a valuable approach in this context. In this paper, we present WeirdFlows, a top-down search pipeline for detecting potentially fraudulent transactions and non-compliant agents. In a transaction network, fraud attempts are often based on complex transaction patterns that change over time to avoid detection. The WeirdFlows pipeline requires neither an a priori set of patterns nor a training set. In addition, by providing elements to explain the anomalies found, it facilitates and supports the work of an AFC analyst. We evaluate WeirdFlows on a dataset from Intesa Sanpaolo (ISP) bank, comprising 80 million cross-country transactions over 15 months, benchmarking our implementation of the algorithm. The results, corroborated by ISP AFC experts, highlight its effectiveness in identifying suspicious transactions and actors, particularly in the context of the economic sanctions imposed in the EU after February 2022. This demonstrates \textit{WeirdFlows}' capability to handle large datasets, detect complex transaction patterns, and provide the necessary interpretability for formal AFC investigations.

cs.CY↗

arXiv2023

Narratives of War: Ukrainian Memetic Warfare on Twitter

The 2022 Russian invasion of Ukraine has seen an intensification in the use of social media by governmental actors in cyber warfare. Wartime communication via memes has been a successful strategy used not only by independent accounts such as @uamemesforces, but also-for the first time in a full-scale interstate war-by official Ukrainian government accounts such as @Ukraine and @DefenceU. We study this prominent example of memetic warfare through the lens of its narratives, and find them to be a key component of success: tweets with a 'victim' narrative garner twice as many retweets. However, malevolent narratives focusing on the enemy resonate more than those about heroism or victims with countries providing more assistance to Ukraine. Our findings present a nuanced examination of Ukraine's influence operations and of the worldwide response to it, thus contributing new insights into the evolution of socio-technical systems in times of war.

cs.CY↗

arXiv2023

The Thin Ideology of Populist Advertising on Facebook during the 2019 EU Elections

Social media has been an important tool in the expansion of the populist message, and it is thought to have contributed to the electoral success of populist parties in the past decade. This study compares how populist parties advertised on Facebook during the 2019 European Parliamentary election. In particular, we examine commonalities and differences in which audiences they reach and on which issues they focus. By using data from Meta (previously Facebook) Ad Library, we analyze 45k ad campaigns by 39 parties, both populist and mainstream, in Germany, United Kingdom, Italy, Spain, and Poland. While populist parties represent just over 20% of the total expenditure on political ads, they account for 40% of the total impressions$\unicode{x2013}$most of which from Eurosceptic and far-right parties$\unicode{x2013}$thus hinting at a competitive advantage for populist parties on Facebook. We further find that ads posted by populist parties are more likely to reach male audiences, and sometimes much older ones. In terms of issues, populist politicians focus on monetary policy, state bureaucracy and reforms, and security, while the focus on EU and Brexit is on par with non-populist, mainstream parties. However, issue preferences are largely country-specific, thus supporting the view in political science that populism is a "thin ideology", that does not have a universal, coherent policy agenda. This study illustrates the usefulness of publicly available advertising data for monitoring the populist outreach to, and engagement with, millions of potential voters, while outlining the limitations of currently available data.

cs.SI↗

arXiv2022

Analyzing and visualizing polarization and balance with signed networks: the U.S. Congress case study

Signed networks and balance theory provide a natural setting for real-world scenarios that show polarization dynamics, positive/negative relationships, and political partisanship. For example, they have been proven effective in studying the increasing polarization of the votes in the two chambers of the U.S. Congress from World War II on. To provide further insights into this particular case study, we propose the application of a pipeline to analyze and visualize a signed graph's configuration based on the exploitation of the corresponding Laplacian matrix' spectral properties. The overall methodology is comparable with others based on the frustration index, but it has at least two main advantages: first, it requires a much lower computational cost; second, it allows for a quantitative and visual assessment of how arbitrarily small subgraphs (even single nodes) contribute to the overall balance (or unbalance) of the network. The proposed pipeline allows the exploration of polarization dynamics shown by the U.S. Congress from 1945 to 2020 at different resolution scales. In fact, we are able to spot and point out the influence of some (groups of) congressmen in the overall balance, as well as to observe and explore polarization's evolution of both chambers across the years.

cs.SI↗

arXiv2021

Clandestino or Rifugiato? Anti-immigration Facebook Ad Targeting in Italy

Monitoring advertising around controversial issues is an important step in ensuring accountability and transparency of political processes. To that end, we use the Facebook Ads Library to collect 2312 migration-related advertising campaigns in Italy over one year. Our pro- and anti-immigration classifier (F1=0.85) reveals a partisan divide among the major Italian political parties, with anti-immigration ads accounting for nearly 15M impressions. Although composing 47.6% of all migration-related ads, anti-immigration ones receive 65.2% of impressions. We estimate that about two thirds of all captured campaigns use some kind of demographic targeting by location, gender, or age. We find sharp divides by age and gender: for instance, anti-immigration ads from major parties are 17% more likely to be seen by a male user than a female. Unlike pro-migration parties, we find that anti-immigration ones reach a similar demographic to their own voters. However their audience change with topic: an ad from anti-immigration parties is 24% more likely to be seen by a male user when the ad speaks about migration, than if it does not. Furthermore, the viewership of such campaigns tends to follow the volume of mainstream news around immigration, supporting the theory that political advertisers try to "ride the wave" of current news. We conclude with policy implications for political communication: since the Facebook Ads Library does not allow to distinguish between advertisers intentions and algorithmic targeting, we argue that more details should be shared by platforms regarding the targeting configuration of socio-political campaigns.

cs.SI↗

arXiv2020

Facebook Ads: Politics of Migration in Italy

Targeted online advertising is on the forefront of political communication, allowing hyper-local advertising campaigns around elections and issues. In this study, we employ a new resource for political ad monitoring -- Facebook Ads Library -- to examine advertising concerning the issue of immigration in Italy. A crucial topic in Italian politics, it has recently been a focus of several populist movements, some of which have adopted social media as a powerful tool for voter engagement. Indeed, we find evidence of targeting by the parties both in terms of geography and demographics (age and gender). For instance, Five Star Movement reaches a younger audience when advertising about immigration, while other parties' ads have a more male audience when advertising on this issue. We also notice a marked rise in advertising volume around elections, as well as a shift to more general audience. Thus, we illustrate political advertising targeting that likely has an impact on public opinion on a topic involving potentially vulnerable populations, and urge the research community to include online advertising in the monitoring of public discourse.

cs.CY↗