Analyzing mass media influence using natural language processing and time series analysis

Albanese, Federico; Pinto, Sebastián; Semeshenko, Viktoriya; Balenzuela, Pablo

doi:10.1088/2632-072x/ab8784

Cited by 14 publications

(11 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The matrix H has a representation of the tweets in the topic space, in which the columns are the degree of membership of each tweet to a given topic. On the other hand, the matrix W provides the combination of terms which describes each topic [42].…”

Section: Natural Language Processing Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Breaking the Communities: Characterizing community changing users using text mining and graph machine learning on Twitter

Albanese¹,

Lombardi²,

Feuerstein³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

The formation of majorities in public discussions often depends on individuals who shift their opinion over time. The detection and characterization of these type of individuals is therefore extremely important for political analysis of social networks. In this paper, we study changes in individual's affiliations on Twitter using natural language processing techniques and graph machine learning algorithms. In particular, we collected 9 million Twitter messages from 1.5 million users and constructed the retweet networks. We identified communities with explicit political orientation and topics of discussion associated to them which provide the topological representation of the political map on Twitter in the analyzed periods. With that data, we present a machine learning framework for social media users classification which efficiently detects "shifting users" (i.e. users that may change their affiliation over time). Moreover, this machine learning framework allows us to identify not only which topics are more persuasive (using low dimensional topic embedding), but also which individuals are more likely to change their affiliation given their topological properties in a Twitter graph.

show abstract

Section: Natural Language Processing Featuresmentioning

confidence: 99%

“…The decomposition dimension was swept between 5 and 30, and for each dataset we chose a number of topics in the corpus so as to have a clear interpretation of each one. The same methodology was used and described in [11,42].…”

Section: Natural Language Processing Featuresmentioning

confidence: 99%

Breaking the Communities: Characterizing community changing users using text mining and graph machine learning on Twitter

Albanese¹,

Lombardi²,

Feuerstein³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another realistic feature at the collective level that was included in opinion models is an external mass media propaganda, which influences public opinion in the preferred direction set by the propaganda [ 25 , 37 , 38 , 39 , 40 , 41 ]. The case scenario of a propaganda that oscillates in time was considered in several studies [ 24 , 25 , 42 , 43 , 44 , 45 , 46 ] where, in analogy with spin systems in statistical physics, the propaganda can be viewed as an external oscillating field that has an influence on the opinion update of every agent.…”

Section: Introductionmentioning

confidence: 99%

Contrarian Voter Model under the Influence of an Oscillating Propaganda: Consensus, Bimodal Behavior and Stochastic Resonance

Giménez

Reinaudi

Vázquez

2022

Entropy

View full text Add to dashboard Cite

We study the contrarian voter model for opinion formation in a society under the influence of an external oscillating propaganda and stochastic noise. Each agent of the population can hold one of two possible opinions on a given issue—against or in favor—and interacts with its neighbors following either an imitation dynamics (voter behavior) or an anti-alignment dynamics (contrarian behavior): each agent adopts the opinion of a random neighbor with a time-dependent probability p(t), or takes the opposite opinion with probability 1−p(t). The imitation probability p(t) is controlled by the social temperature T, and varies in time according to a periodic field that mimics the influence of an external propaganda, so that a voter is more prone to adopt an opinion aligned with the field. We simulate the model in complete graph and in lattices, and find that the system exhibits a rich variety of behaviors as T is varied: opinion consensus for T=0, a bimodal behavior for T<Tc, an oscillatory behavior where the mean opinion oscillates in time with the field for T>Tc, and full disorder for T≫1. The transition temperature Tc vanishes with the population size N as Tc≃2/lnN in complete graph. In addition, the distribution of residence times tr in the bimodal phase decays approximately as tr−3/2. Within the oscillatory regime, we find a stochastic resonance-like phenomenon at a given temperature T*. Furthermore, mean-field analytical results show that the opinion oscillations reach a maximum amplitude at an intermediate temperature, and that exhibit a lag with respect to the field that decreases with T.

show abstract

“…Researchers in the field chiefly address the question of which individuals should be influenced-under a budget constraint-to achieve maximum spread. Studies on IM allow us to better understand different aspects of influencing attempts, such as under which conditions one can effectively introduce the desired opinion in a population [12,13], how influencing strategies interact when competition for influence takes place [14,15], or what the effect of external influence is on reaching consensus in a population [16].…”

Section: Introductionmentioning

confidence: 99%

Shadowing and shielding: Effective heuristics for continuous influence maximisation in the voting dynamics

2021

View full text Add to dashboard Cite

Influence maximisation, or how to affect the intrinsic opinion dynamics of a social group, is relevant for many applications, such as information campaigns, political competition, or marketing. Previous literature on influence maximisation has mostly explored discrete allocations of influence, i.e. optimally choosing a finite fixed number of nodes to target. Here, we study the generalised problem of continuous influence maximisation where nodes can be targeted with flexible intensity. We focus on optimal influence allocations against a passive opponent and compare the structure of the solutions in the continuous and discrete regimes. We find that, whereas hub allocations play a central role in explaining optimal allocations in the discrete regime, their explanatory power is strongly reduced in the continuous regime. Instead, we find that optimal continuous strategies are very well described by two other patterns: (i) targeting the same nodes as the opponent (shadowing) and (ii) targeting direct neighbours of the opponent (shielding). Finally, we investigate the game-theoretic scenario of two active opponents and show that the unique pure Nash equilibrium is to target all nodes equally. These results expose fundamental differences in the solutions to discrete and continuous regimes and provide novel effective heuristics for continuous influence maximisation.

show abstract

Analyzing mass media influence using natural language processing and time series analysis

Cited by 14 publications

References 32 publications

Breaking the Communities: Characterizing community changing users using text mining and graph machine learning on Twitter

Breaking the Communities: Characterizing community changing users using text mining and graph machine learning on Twitter

Contrarian Voter Model under the Influence of an Oscillating Propaganda: Consensus, Bimodal Behavior and Stochastic Resonance

Shadowing and shielding: Effective heuristics for continuous influence maximisation in the voting dynamics

Contact Info

Product

Resources

About