Conformity: A Path-Aware Homophily Measure for Node-Attributed Networks

Rossetti, Giulio; Citraro, Salvatore; Milli, Letizia

doi:10.1109/mis.2021.3051291

Cited by 17 publications

(28 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to address this point we considered not only emotional complexity but also emotional entropy, the latter being a non-network metric considering shifts between emotional states in the whole datasets. Future research could integrate emotional complexity and entropy with recent measures like conformity [57], mixing network structure and node features.…”

Section: Discussionmentioning

confidence: 99%

Cognitive network science as a framework for detecting structural patterns and emotions in suicide letters

Stella¹,

Swanson²,

Y³

et al. 2021

Preprint

View full text Add to dashboard Cite

Communicating one's mindset means transmitting complex relationships between concepts and emotions. Using cognitive network science, we reconstruct the mindset around suicide as communicated in 139 genuine suicide notes. Despite their negative context, suicide notes are surprisingly positively valenced and their ending statements are markedly more emotional, i.e. elicit deeper fear/sadness but also stronger joy/trust and anticipation, than their main body. By using emotional states from the Emotional Recall Task, we "open the lid" of suicidal narratives and compare their emotional backbone against emotion recall in mentally healthy individuals. Supported by psychological literature, we introduce emotional complexity as an affective analogue of structural balance theory, measuring how elementary cycles (closed triads) of emotion co-occurrences mix positive, negative and neutral states in narratives and recollections. Both authors of suicide notes and healthy individuals exhibit less complexity and more emotional coherence than expected by chance. However, suicide narratives display higher complexity, i.e. a lower level of coherently valenced triads, than healthy individuals recalling the same states. Entropy measures identified a similar tendency for suicide letters to shift more frequently between contrasting emotional states. Our results demonstrate that suicide notes possess highly contrastive narratives of emotions, more complex than expected by null models and healthy populations.

show abstract

Section: Discussionmentioning

confidence: 99%

Cognitive network science as a framework for detecting structural patterns and emotions in suicide letters

Stella¹,

Swanson²,

Y³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In [28], Conformity by gender was measured in a daily-aggregated static network, where males were on average more assortative than females. We aim to provide more insights into students' daily routines, e.g., whether mixing patterns differ by day and night or by weekdays and weekends.…”

Section: Methodsmentioning

confidence: 99%

“…We describe ∆-Conformity as a conservative extension of Conformity (previously introduced in [28]) to Feature-rich Stream Graph.…”

Section: Multiscale Homophily In Feature-rich Stream Graphmentioning

confidence: 99%

See 1 more Smart Citation

Δ-Conformity: Multi-scale Node Assortativity in Feature-rich Stream Graphs

Citraro¹,

Milli²,

Cazabet³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Heterogeneity is a key aspect of complex networks, often emerging by looking at the distribution of node properties, from the milestone observations on the degree to the recent developments in mixing pattern estimation. Mixing patterns, in particular, refer to nodes' connectivity preferences with respect to an attribute label. Social networks are mostly characterized by assortative/homophilic behaviour, where nodes are more likely to be connected with similar ones. Recently, assortative mixing is increasingly measured in a multi-scale fashion to overcome well-known limitations of classic scores. Such multi-scale strategies can capture heterogeneous behaviors among node homophily, but they ignore an important, often available, addendum in real-world systems: the time when edges are present and the time-varying paths they form accordingly. Hence, temporal homophily is still little understood in complex networks. In this work we aim to cover this gap by introducing the ∆-Conformity measure, a multi-scale, path-aware, node homophily estimator within the new framework of feature-rich stream graphs. A rich experimental section analyzes ∆-Conformity trends over time, spanning the analysis from real-life social interaction networks to a specific case-study about the Bitcoin Transaction Network.

show abstract

“…Since H and H exp do not take the homophilic contribution of each community/node into account explicitly, we also provide (i) a function capturing noise within communities (i.e., the percentage of the majority attribute value within a cluster), namely Purity (Citraro and Rossetti 2019), and (ii) two measures for explaining the homophilic contribution of each node, namely Peel's assortativity (Peel et al 2018) and Conformity (Rossetti et al 2020). Given a community C, its purity P c is the product of the frequencies of the most frequent categorical attribute values carried by the nodes within C, formally:…”

Section: X-mark Characterizationmentioning

confidence: 99%

X-Mark: a benchmark for node-attributed community discovery algorithms

Citraro

Rossetti

2021

Soc. Netw. Anal. Min.

Self Cite

View full text Add to dashboard Cite

Grouping well-connected nodes that also result in label-homogeneous clusters is a task often known as attribute-aware community discovery. While approaching node-enriched graph clustering methods, rigorous tools need to be developed for evaluating the quality of the resulting partitions. In this work, we present X-Mark, a model that generates synthetic node-attributed graphs with planted communities. Its novelty consists in forming communities and node labels contextually while handling categorical or continuous attributive information. Moreover, we propose a comparison between attribute-aware algorithms, testing them against our benchmark. Accordingly to different classification schema from recent state-of-the-art surveys, our results suggest that X-Mark can shed light on the differences between several families of algorithms.

show abstract

Conformity: A Path-Aware Homophily Measure for Node-Attributed Networks

Cited by 17 publications

References 25 publications

Cognitive network science as a framework for detecting structural patterns and emotions in suicide letters

Cognitive network science as a framework for detecting structural patterns and emotions in suicide letters

Δ-Conformity: Multi-scale Node Assortativity in Feature-rich Stream Graphs

X-Mark: a benchmark for node-attributed community discovery algorithms

Contact Info

Product

Resources

About