Inferred Networks and the Social Determinants of Health

Sanjel, Prashant; Matta, John

doi:10.1007/978-3-030-93413-2_58

Cited by 2 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We focus on these clusterings for further visualization and analysis. Both here and in [16], the NBR-Clust method produced some of the best results. It is stated in [17] that node-removal clustering methods such as NBR-Clust are effective at identifying noise, creating cleaner clusters than other methods.…”

Section: Resultsmentioning

confidence: 67%

“…solving the lasso problem for each variable using coordinate descent [43]. Both glasso and MB inference methods were chosen because of previous empirical evidence showing good results for methods that utilize lasso regression [16].…”

Section: Plos Onementioning

confidence: 99%

“…This paper is an extension of [16], in which the methodology presented in [42] was followed. We continue to employ this methodology, as the purpose of this paper is to examine its use when mining larger, similar medical data sets, with networks inferred using different…”

Section: Cluster Evaluationmentioning

confidence: 99%

“…algorithms. The SATHCAP dataset is good for this examination, as it includes variables specifically for HIV as well as other attributes present in the dataset used in [16].…”

Section: Plos Onementioning

confidence: 99%

“…Alcohol and drug use can be seen affecting the living conditions of members in cluster 0, with 22.5% currently living in a shelter, boarding house, or halfway house. It is interesting that in [16] HIV status is differentiated mainly by the prior occurrence of STDs. The SATHCAP dataset is highly focused on drug use, which is evident in this clustering.…”

Section: Hiv Statusmentioning

confidence: 99%

See 4 more Smart Citations

Inferred networks, machine learning, and health data

et al. 2023

Self Cite

View full text Add to dashboard Cite

This paper presents a network science approach to investigate a health information dataset, the Sexual Acquisition and Transmission of HIV Cooperative Agreement Program (SATHCAP), to uncover hidden relationships that can be used to suggest targeted health interventions. From the data, four key target variables are chosen: HIV status, injecting drug use, homelessness, and insurance status. These target variables are converted to a graph format using four separate graph inference techniques: graphical lasso, Meinshausen Bühlmann (MB), k-Nearest Neighbors (kNN), and correlation thresholding (CT). The graphs are then clustered using four clustering methods: Louvain, Leiden, and NBR-Clust with VAT and integrity. Promising clusters are chosen using internal evaluation measures and are visualized and analyzed to identify marker attributes and key relationships. The kNN and CT inference methods are shown to give useful results when combined with NBR-Clust clustering. Examples of cluster analysis indicate that the methodology produces results that will be relevant to the public health community.

show abstract

Section: Resultsmentioning

confidence: 67%

Section: Plos Onementioning

confidence: 99%

Section: Cluster Evaluationmentioning

confidence: 99%

“…algorithms. The SATHCAP dataset is good for this examination, as it includes variables specifically for HIV as well as other attributes present in the dataset used in [16].…”

Section: Plos Onementioning

confidence: 99%

Section: Hiv Statusmentioning

confidence: 99%

See 3 more Smart Citations