Cognitive Network Science for Understanding Online Social Cognitions: A Brief Review

Stella, Massimo

doi:10.1111/tops.12551

Cited by 33 publications

(80 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Understanding the language used in online social networks is a crucial task with plenty of implications, especially in the case of automated accounts: in fact, a detailed characterization of the language used by social bots permits to study how automated accounts influence humans [ 19 , 58 ]. Even if a detailed analysis of the language used by the different categories and its implications is beyond the scope of the present paper, we present some brief insights about the frequencies of hashtags used by the various user groups.…”

Section: Resultsmentioning

confidence: 99%

Brexit and bots: characterizing the behaviour of automated accounts on Twitter during the UK election

Lambiotte

Saracco

2022

EPJ Data Sci.

View full text Add to dashboard Cite

Online Social Networks (OSNs) offer new means for political communications that have quickly begun to play crucial roles in political campaigns, due to their pervasiveness and communication speed. However, the OSN environment is quite slippery and hides potential risks: many studies presented evidence about the presence of d/misinformation campaigns and malicious activities by genuine or automated users, putting at severe risk the efficiency of online and offline political campaigns. This phenomenon is particularly evident during crucial political events, as political elections. In the present paper, we provide a comprehensive description of the networks of interactions among users and bots during the UK elections of 2019. In particular, we focus on the polarised discussion about Brexit on Twitter, analysing a data set made of more than 10 millions tweets posted for over a month. We found that the presence of automated accounts infected the debate particularly in the days before the UK national elections, in which we find a steep increase of bots in the discussion; in the days after the election day, their incidence returned to values similar to the ones observed few weeks before the elections. On the other hand, we found that the number of suspended users (i.e. accounts that were removed by the platform for some violation of the Twitter policy) remained constant until the election day, after which it reached significantly higher values. Remarkably, after the TV debate between Boris Johnson and Jeremy Corbyn, we observed the injection of a large number of novel bots whose behaviour is markedly different from that of pre-existing ones. Finally, we explored the bots’ political orientation, finding that their activity is spread across the whole political spectrum, although in different proportions, and we studied the different usage of hashtags and URLs by automated accounts and suspended users, targeting the formation of common narratives in different sides of the debate.

show abstract

Section: Resultsmentioning

confidence: 99%

Brexit and bots: characterizing the behaviour of automated accounts on Twitter during the UK election

Lambiotte

Saracco

2022

EPJ Data Sci.

View full text Add to dashboard Cite

show abstract

“…Similarly, DASentimental could be applied to track the change of explicit expression of depression, anxiety, and stress over history, quantified through the emotions of modern individuals. This would highlight changes in norms towards emotional expression and historical events (e.g., "pandemic"), thus complementing other recent approaches in cognitive network science [9,30,[59][60][61] and sentiment/emotional profiling [51,55,62,63] by bringing to the table a quantitative, automatic quantification of depression, anxiety, and stress in texts.…”

Section: Limitations and Future Researchmentioning

confidence: 85%

“…This has two implications: First, a stronger model might use distances and valences when focusing only on fluency tasks. Second, for text analysis and even cognitive social media mining [55], DASentimental's machine learning pipeline could be used to detect any kind of target emotion (e.g., "surprise" or "love"). Furthermore, DASentimenal could be used on future datasets relating DAS levels with ERT data and demographics to explore how age, gender, or physical health can influence depression, anxiety, and stress detection.…”

Section: Limitations and Future Researchmentioning

confidence: 99%

DASentimental: Detecting Depression, Anxiety, and Stress in Texts via Emotional Recall, Cognitive Networks, and Machine Learning

Fatima

Hills

et al. 2021

BDCC

Self Cite

View full text Add to dashboard Cite

Most current affect scales and sentiment analysis on written text focus on quantifying valence/sentiment, the primary dimension of emotion. Distinguishing broader, more complex negative emotions of similar valence is key to evaluating mental health. We propose a semi-supervised machine learning model, DASentimental, to extract depression, anxiety, and stress from written text. We trained DASentimental to identify how N = 200 sequences of recalled emotional words correlate with recallers’ depression, anxiety, and stress from the Depression Anxiety Stress Scale (DASS-21). Using cognitive network science, we modeled every recall list as a bag-of-words (BOW) vector and as a walk over a network representation of semantic memory—in this case, free associations. This weights BOW entries according to their centrality (degree) in semantic memory and informs recalls using semantic network distances, thus embedding recalls in a cognitive representation. This embedding translated into state-of-the-art, cross-validated predictions for depression (R = 0.7), anxiety (R = 0.44), and stress (R = 0.52), equivalent to previous results employing additional human data. Powered by a multilayer perceptron neural network, DASentimental opens the door to probing the semantic organizations of emotional distress. We found that semantic distances between recalls (i.e., walk coverage), was key for estimating depression levels but redundant for anxiety and stress levels. Semantic distances from “fear” boosted anxiety predictions but were redundant when the “sad–happy” dyad was considered. We applied DASentimental to a clinical dataset of 142 suicide notes and found that the predicted depression and anxiety levels (high/low) corresponded to differences in valence and arousal as expected from a circumplex model of affect. We discuss key directions for future research enabled by artificial intelligence detecting stress, anxiety, and depression in texts.

show abstract

“…In the burgeoning field of socio-semantic analysis of digital exchanges, several other “variables” have been considered and have shown some explanatory power for the link between the social and the semantic. For instance, expert and contextual knowledge 1 , 55 , emotions 16 , actors’ characteristics such as personality, roles or status 56 – 59 have been considered. These considerations could be reinterpreted under the lens of a mutualistic framework with novel interrogations around factors affecting the grouping of actors above and beyond the semantic and factors affecting the primacy of certain topics.…”

Section: Discussionmentioning

confidence: 99%

Socio-semantic networks as mutualistic networks

St-Onge

Renaud-Desjardins²,

Mongeau

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Several studies have shown that discourse and social relationships are intertwined and co-evolve. However, we lack theoretical models to explain the phenomenon. Inspired by recent work in ecology, we propose to model socio-semantic networks as an interaction between two intermingled data generating processes: a social community process and a document-based process. We consider the link between semantic and social ties as analogous to the interactions found in pollination networks whereby agents visit hidden topics in a similar way that insects visit specific plants for pollination. We use the ENRON socio-semantic email network to investigate if it exhibits properties that characterize mutualistic networks, namely moderate connectance, heterogeneous degree distribution, moderate modularity and high nestedness. To do so, we build a plant-pollinator matrix where “insect species” are communities detected via block modelling, “plant species” are latent topics detected with topic modelling, and the interaction between the two is the total number of visits a community makes to specific topics. Our results show that the ENRON socio-semantic interaction matrix respects the aforementioned criteria of mutualism paving the way for the development of a relevant framework to better understand the dynamic of human socio-semantic interactions.

show abstract

Cognitive Network Science for Understanding Online Social Cognitions: A Brief Review

Cited by 33 publications

References 90 publications

Brexit and bots: characterizing the behaviour of automated accounts on Twitter during the UK election

Brexit and bots: characterizing the behaviour of automated accounts on Twitter during the UK election

DASentimental: Detecting Depression, Anxiety, and Stress in Texts via Emotional Recall, Cognitive Networks, and Machine Learning

Socio-semantic networks as mutualistic networks

Contact Info

Product

Resources

About