Affinity propagation: An exemplar-based tool for clustering in psychological research

Population Monitoring of the Red Junglefowl Based on Acoustic Signal Recognition Technology

Vocalisation is a crucial means of communication for birds and plays a key role in survival and reproductive success. Individual differences in songs have been used for identification in many animals, but few studies have integrated song individuality into wildlife population monitoring. The male red junglefowl Gallus gallus jabouillei is a tropical forest bird that primarily uses acoustic signals for conspecific communication. From July to August 2020, the calls of 34 pasture-raised red junglefowl were recorded for individual identification based on vocalisations. Fieldwork was conducted from March to May 2021 in the Datian National Nature Reserve, Hainan, China, during which microphone arrays were deployed to record the calls of wild red junglefowl throughout their breeding season. Discriminant function analysis (DFA) was applied to identify pasture-raised red junglefowl individuals, achieving a correct identification rate of 95.7%. Affinity propagation (AP) clustering was used to perform unsupervised clustering based on pairwise syllable similarities, resulting in 34 clusters corresponding to the actual number of individuals, with a correct syllable type recognition rate of 99.4%. Kaleidoscope software was used to extract the call during the breeding period of the wild population of red junglefowl; the precision rate was 80.38%, and the recall rate was 75.85%. Using AP clustering for vocalisation analysis, the estimated population in the core area was approximately 205 male individuals, with a manual verification accuracy of 82.5%. This result is slightly lower than the estimate of 234 individuals obtained using vocal count and random encounter methods. Our study demonstrated the potential of affinity propagation clustering techniques for estimating the population size of wild red junglefowl.

The molecular subtypes of autoimmune diseases

Autoimmune diseases (ADs) are characterized by their complexity and a wide range of clinical differences. Despite patients presenting with similar symptoms and disease patterns, their reactions to treatments may vary. The current approach of personalized medicine, which relies on molecular data, is seen as an effective method to address the variability in these diseases. This review examined the pathologic classification of ADs, such as multiple sclerosis and lupus nephritis, over time. Acknowledging the limitations inherent in pathologic classification, the focus shifted to molecular classification to achieve a deeper insight into disease heterogeneity. The study outlined the established methods and findings from the molecular classification of ADs, categorizing systemic lupus erythematosus (SLE) into four subtypes, inflammatory bowel disease (IBD) into two, rheumatoid arthritis (RA) into three, and multiple sclerosis (MS) into a single subtype. It was observed that the high inflammation subtype of IBD, the RA inflammation subtype, and the MS "inflammation & EGF" subtype share similarities. These subtypes all display a consistent pattern of inflammation that is primarily driven by the activation of the JAK-STAT pathway, with the effective drugs being those that target this signaling pathway. Additionally, by identifying markers that are uniquely associated with the various subtypes within the same disease, the study was able to describe the differences between subtypes in detail. The findings are expected to contribute to the development of personalized treatment plans for patients and establish a strong basis for tailored approaches to treating autoimmune diseases.

Cluster Analysis Method Reveals Gender Attitudes in Sociosexual Orientation of a Southern Italy Population During the COVID-19 Lockdown

Introduction

The COVID-19 epidemic and its lockdown dramatically impacted the general well-being of the population and affected sociosexual experiences, thus modifying sexual behavior, desire, and well-being. Clustering analysis has not yet been applied to research and data investigating sociosexuality. The cluster analysis method could be a valid support for clinicians in investigating the condition of a population with respect to problems related to sociosexuality. The aim of the present study was to analyze the different perceptions of the sociosexual experiences in southern population during the COVID-19 pandemic.

Methods

We enrolled 734 (450 female) participants with a carried out anonymous web-based survey from the 16th of April 2020 to the 3rd June of 2020. The revised Sociosexual Orientation Inventory (SOI-R) is a self-report test assessing three theoretically meaningful facets of sociosexual orientation (behavior, attitude, and desire).

Results

We found eleven clusters, and the findings showed, for the first time, an intra- and inter-diagnostic heterogeneity in the sexual profile of participants. Theoretically, we identified subtype clusters whose sexual attitude was to avoid sexual promiscuity with significant gender differences. Women show a greater propensity for attitude and desire facet than men.

Conclusions

Our new method of unsupervised learning could represent a reliable tool to support socio-cultural analysis studies on issues influenced by cultural mechanisms in a quick and explanatory way, as in the case of sexual orientation and attitude differences between men and women.

Social and Policy Implications

Understanding these gaps is fundamental for policy makers, managers of social networks, those who deal with engaged couples and families, and sexuality starting from the very youngest adolescents. We claim to devise a strategy to measure how much a sexist culture implicitly and explicitly limits the freedom of sexual expression and how this can affect psycho-sexual well-being in a society.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13178-022-00771-2.

A Novel Approach for Cognitive Clustering of Parkinsonisms through Affinity Propagation

Cluster analysis is widely applied in the neuropsychological field for exploring patterns in cognitive profiles, but traditional hierarchical and non-hierarchical approaches could be often poorly effective or even inapplicable on certain type of data. Moreover, these traditional approaches need the initial specification of the number of clusters, based on a priori knowledge not always owned. For this reason, we proposed a novel method for cognitive clustering through the affinity propagation (AP) algorithm. In particular, we applied the AP clustering on the regression residuals of the Mini Mental State Examination scores—a commonly used screening tool for cognitive impairment—of a cohort of 49 Parkinson’s disease, 48 Progressive Supranuclear Palsy and 44 healthy control participants. We found four clusters, where two clusters (68 and 30 participants) showed almost intact cognitive performance, one cluster had a moderate cognitive impairment (34 participants), and the last cluster had a more extensive cognitive deficit (8 participants). The findings showed, for the first time, an intra- and inter-diagnostic heterogeneity in the cognitive profile of Parkinsonisms patients. Our novel method of unsupervised learning could represent a reliable tool for supporting the neuropsychologists in understanding the natural structure of the cognitive performance in the neurodegenerative diseases.

A Method for Estimating Driving Factors of Illicit Trade Using Node Embeddings and Clustering

The trade on illegal goods and services, also known as illicit trade, is expected to drain 4.2 trillion dollars from the world economy and put 5.4 million jobs at risk by 2022. These estimates reflect the importance of combating illicit trade, as it poses a danger to individuals and undermines governments. To do so, however, we have to first understand the factors that influence this type of trade. Therefore, we present in this article a method that uses node embeddings and clustering to compare a country based illicit supply network to other networks that represent other types of country relationships (e.g., free trade agreements, language). The results offer initial clues on the factors that might be driving the illicit trade between countries.