Membership Model Research Articles

Flexible regularized estimation in high-dimensional mixed membership models

Mixed membership models are an extension of finite mixture models, where each observation can partially belong to more than one mixture component. A probabilistic framework for mixed membership models of high-dimensional continuous data is proposed with a focus on scalability and interpretability. The novel probabilistic representation of mixed membership is based on convex combinations of dependent multivariate Gaussian random vectors. In this setting, scalability is ensured through approximations of a tensor covariance structure through multivariate eigen-approximations with adaptive regularization imposed through shrinkage priors. Conditional weak posterior consistency is established on an unconstrained model, allowing for a simple posterior sampling scheme while keeping many of the desired theoretical properties of our model. The model is motivated by two biomedical case studies: a case study on functional brain imaging of children with autism spectrum disorder (ASD) and a case study on gene expression data from breast cancer tissue. These applications highlight how the typical assumption made in cluster analysis, that each observation comes from one homogeneous subgroup, may often be restrictive in several applications, leading to unnatural interpretations of data features.

SEDMA: Self-Distillation with Model Aggregation for Membership Privacy

Membership inference attacks (MIAs) are important measures to evaluate potential risks of privacy leakage from machine learning (ML) models. State-of-the-art MIA defenses have achieved favorable privacy-utility trade-offs using knowledge distillation on split training datasets. However, such defenses increase computational costs as a large number of the ML models must be trained on the split datasets. In this study, we proposed a new MIA defense, called SEDMA, based on self-distillation using model aggregation to mitigate the MIAs, inspired by the model parameter averaging as used in federated learning. The key idea of SEDMA is to split the training dataset into several parts and aggregate multiple ML models trained on each split for self-distillation. The intuitive explanation of SEDMA is that model aggregation prevents model over-fitting by smoothing information related to the training data among the multiple ML models and preserving the model utility, such as in federated learning. Through our experiments on major benchmark datasets (Purchase100, Texas100, and CIFAR100), we show that SEDMA outperforms state-of-the-art MIA defenses in terms of membership privacy (MIA accuracy), model accuracy, and computational costs. Specifically, SEDMA incurs at most approximately 3 - 5% model accuracy drop, while achieving the lowest MIA accuracy in state-of-the-art empirical MIA defenses. For computational costs, SEDMA takes significantly less processing time than a defense with the state-of-the-art privacy-utility trade-offs in previous defenses. SEDMA achieves both favorable privacy-utility trade-offs and low computational costs.

Modeling non-compensatory strategies on path choices in a complex urban rail transit network considering characteristics of transfer passengers and trips

The path choice behavior of passengers is highly affected by transfer across different lines in a complex urban rail transit network that provides passengers with multiple path alternatives. Hence, the design of a transfer station in terms of its layout and transfer paths is gaining impact on passengers’ path choices in an expanding urban rail transit network. A comprehensive understanding of passenger behavior is essential for transportation planning and management to better design transfer stations and network operations. A questionnaire survey is conducted for passengers’ path choices and perception of transfer stations. Based on the collected data, this study develops a joint path choice model that combines Multinomial Logit Model (MNL) and Constrained Multinomial Logit Model (CMNL) to capture both compensatory and non-compensatory behavior, especially the latter one, which has not been fully explored in the current literature. Further, a membership model is constructed to analyze the heterogeneity of passengers’ non-compensatory behavior on transfer walking time. Four behavioral segments are identified in regard to passenger and trip characteristics. The estimation results show that the proposed model outperforms the general MNL model and allows a more convincing explanation of urban rail transit passenger behavior. Several research findings are derived, including that transfer walking time is more negatively perceived by passengers without non-compensatory behavior. Moreover, in-vehicle travel time is likely to be a minor consideration for passengers with non-compensatory behavior. The proposed model and framework enable urban rail transit operators and planners to retrieve more precise prediction of passenger flows and enhance service quality.

Modeling Historic Arsenic Exposures and Spatial Risk for Bladder Cancer.

Arsenic is a bladder carcinogen though less is known regarding the specific temporal relationship between exposure and bladder cancer diagnosis. In this study, we modeled time-varying mixtures of arsenic exposures at many historic temporal windows to evaluate their association with bladder cancer risk in the New England Bladder Cancer Study. We used arsenic exposure estimates up to 60 years prior to study entry and compared the goodness of fit of models using these mixtures to those using summary measures of arsenic exposures. We used the Bayesian index low rank kriging multiple membership model (LRK-MMM) to estimate the associations of these mixtures with bladder cancer and estimate cumulative spatial risk for bladder cancer using participants' residential histories. We found consistent evidence that modeling arsenic exposures as a time-varying mixture provided better fit to the data than using a single arsenic exposure summary measure. We estimated several positive though not significant associations of the time-varying arsenic mixtures with bladder cancer having odds ratios (ORs) of 1.03-1.14 and identified many significant and positive associations for an interaction among those who consumed water from a private dug well (ORs 1.28-1.60). Arsenic exposures 40-50 years before study entry received elevated importance weights in these mixtures. Additionally, we found two small areas of elevated cumulative spatial risk for bladder cancer in southern New Hampshire and in south central Maine. These results emphasize the importance of considering time-varying mixtures of exposures for diseases with long latencies such as bladder cancer.

Sub-communities of the vaginal microbiota in pregnant and non-pregnant women.

Diverse and non-Lactobacillus-dominated vaginal microbial communities are associated with adverse health outcomes such as preterm birth and the acquisition of sexually transmitted infections. Despite the importance of recognizing and understanding the key risk-associated features of these communities, their heterogeneous structure and properties remain ill-defined. Clustering approaches are commonly used to characterize vaginal communities, but they lack sensitivity and robustness in resolving substructures and revealing transitions between potential sub-communities. Here, we address this need with an approach based on mixed membership topic models. Using longitudinal data from cohorts of pregnant and non-pregnant study participants, we show that topic models more accurately describe sample composition, longitudinal changes, and better predict the loss of Lactobacillus dominance. We identify several non-Lactobacillus-dominated sub-communities common to both cohorts and independent of reproductive status. In non-pregnant individuals, we find that the menstrual cycle modulates transitions between and within sub-communities, as well as the concentrations of half of the cytokines and 18% of metabolites. Overall, our analyses based on mixed membership models reveal substructures of vaginal ecosystems which may have important clinical and biological associations.

Peer-based research funding as a model for journalism funding

Financing high-quality journalistic reporting is becoming increasingly difficult worldwide and economic pressure has intensified in the wake of the COVID-19 pandemic. While numerous alternative funding possibilities are discussed, ranging from membership models to government funding, they should not compromise the highest possible independence of journalism – a premise that also applies to scientific research. Here, the state is involved in funding, but peer review models reduce funding bias. However, systematic approaches as to how established funding models in research could be transferred to journalism are lacking. We attempt such a systematic transfer using the example of the German Research Foundation ( Deutsche Forschungsgemeinschaft, DFG). The transfer, based on an analysis of the complex DFG funding structures, was validated in 10 interviews with experts from science, journalism and foundations. Building on this, we developed a concept for a German Journalism Foundation ( Deutsche Journalismus-gemeinschaft, DJG), which awards funding to journalists and cooperative projects based on a peer review process. The funding priorities of the proposed organization range from infrastructure support to grants for investigative skills. Thus, unlike other models, it does not focus on funding specific topics in media coverage, but on areas such as innovation support, technology implementation and training. Although the model was designed for Germany, such a systematic transfer could also be tested for other countries.

User preference and willingness-to-pay for operation strategies that enhance safety and convenience of E-scooter sharing services

An electric scooter sharing (ESS) service offers improved convenience by providing accessibility to public transportation stations and mobility for short-distance trips. However, with the growing global popularity of ESS services, concerns related to safety and accidents have also increased. Previous studies explored the factors impacting preferences for ESS services; however, few have considered ways to enhance safety and convenience for users. Therefore, this study investigated the factors that influence people's choice of ESS service for enhancing safety and convenience. Data was collected in Seoul, Korea, and a latent class model was developed, allowing us to estimate the willingness-to-pay (WTP) of various attributes. The estimation results revealed that the “Exclusive e-scooter lane” attribute was the most preferred, followed by “provision of real-time driving information.” The WTP for “Exclusive e-scooter lane” was 47.1% above the base fare. In the estimation results of the class membership model, the preferred attributes were different depending on the class characteristics. The study further revealed variations in WTP for safety and convenience among users and non-users, as well as among non-users depending on their intent to use the ESS service. Furthermore, the proposed attributes constituted factors that affect the safety and convenience of ESS users, and that users were willing to pay an additional cost for them.

The economic hierarchy, linkage, and directions between the three northeastern provinces of China and Russian Far East

AbstractThe economic hierarchy index, economic linkage model, and economic membership model are used to study the economic hierarchy, linkage, and direction of Northeast China and Russian Far East from 2005 to 2020, using the three northeastern provinces of China and 11 Far East federal subjects of Russia as the research objects. The findings are as follows: (1) There is a major polarization effect between developed affluent economic regions and stagnant isolated backward regions in the economic rankings in the 11 Russian Far East federal subjects from 2005 to 2020. (2) The three northeastern provinces of China have more economic linkages with their near Russian neighbors— the east‐oriented Primorsky Territory, the north‐oriented Amur Region, and the north‐oriented Khabarovsk Territory, than with their farther neighbors— the Chukotka Autonomous Area, Magadan Region, and Kamchatka Territory. (3) From 2005 to 2020, the Primorsky Territory has been serving as the focal point of economic linkages between Russian Far East and Northeast China, showing the expanding outward pattern resembling the concentric circles.

Spatial patterns and drivers of native plant diversity in Hainan, China

AbstractHainan Island has the most extensive and well‐preserved tropical forests in China. With rapid economic development of Hainan, biodiversity is increasingly at risk. Determining the spatial patterns of plant diversity in Hainan and explaining the drivers behind plant diversity are important considerations in assessing and maximizing the effectiveness of national parks, such as the newly designated Hainan Rainforest National Park. We assessed phylogenetic diversity patterns, and species richness using 106,252 georeferenced specimen records and a molecular phylogeny of 3,792 native plant species. Based on phylogenetic range‐weighted turnover metrics, we divided Hainan flora into four major floristic units. The Grade of Membership model was used to further verify the four units, and to understand their boundaries and the internal structure of each floristic unit. Finally, the best combination model was used to explore the driving mechanisms underlying the division. Our results reveal that central Hainan is the most important hotspot for plant endemism and diversity, followed by the southern area. Environmental energy is the main factor determining the spatial patterns of native plant diversity on the island, and accessibility has the greatest impact on native plant diversity among social factors. We explore patterns of spatial phylogenetics and biogeography to identify potential priorities for management and conservation drivers of plant diversity patterns across Hainan, to provide the basis for the effective protection of native plant diversity and the improvement of national parks of Hainan Island.

Relating the One-Parameter Logistic Diagnostic Classification Model to the Rasch Model and One-Parameter Logistic Mixed, Partial, and Probabilistic Membership Diagnostic Classification Models

Diagnostic classification models (DCMs) are statistical models with discrete latent variables (so-called skills) to analyze multiple binary variables (i.e., items). The one-parameter logistic diagnostic classification model (1PLDCM) is a DCM with one skill and shares desirable measurement properties with the Rasch model. This article shows that the 1PLDCM is indeed a latent class Rasch model. Furthermore, the relationship of the 1PLDCM to extensions of the DCM to mixed, partial, and probabilistic memberships is treated. It is argued that the partial and probabilistic membership models are also equivalent to the Rasch model. The fit of the different models was empirically investigated using six datasets. It turned out for these datasets that the 1PLDCM always had a worse fit than the Rasch model and mixed and partial membership extensions of the DCM.

How did organ donation in Israel become a club membership model? From civic to communal solidarity in organ sharing

Figuring out what pushes individuals to become organ donors has become the holy grail of social scientists interested in transplantations. In this paper I concentrate on solidarity as a determinant of organ donation and examine it through the history of organ donation in Israel. By following the history of transplantation policies since 1968 and examining them in relation to different types of solidarities, this paper leads to a nuanced understanding of the ties between solidarity and health policy. Attempts to foster an all-encompassing consensus on the definition of brain death yielded the Transplantation and the Brain-Respiratory Death Laws of 2008. It was hoped that a wide “civic solidarity” would render Israel self-sufficient in its organ economy. However, the failure of the law led to the breakdown of civic solidarity in organ donation. As a result, initiatives such as the priority policy and non-directed living organ donations, developed out of a narrower conception of solidarity. Juxtaposing these initiatives sheds light on macro level processes for policy makers and suggests solidarity as a key bioethical concept to understand organ donation policies.

Windstorm Hazardous Area Mapping Using Multi-Criteria Evaluation Techniques of Fuzzy Logic Approach

The goal of this study to determine the most exposed area to windstorm event using fuzzy logic approach. Respond to the Sustainable Development Goal (SDG) 11 and 13, Sustainable Cities and Communities and Climate Action, this study was conducted to establish the preparedness in high-risk region. To achieve this goal, the influencing indicator was identified based on previous study. The related indicator comprises of wind speed, temperature, relative humidity, precipitation, slope, population, road, and land use. Fuzzy membership model implemented to aggregating the indicator to standard value 0 to 1. A fuzzy model was used as a method to integrate the constructed index values, and a frequency ratio was used to determine the fuzzy membership value. Different fuzzy operators (AND, OR, SUM, PRODUCT, and GAMMA) are applied to generate windstorm hazard map. The fuzzy gamma (γ) operator, with γ equal to 0.9, was used to create wind hazard map into different zones of five relative sensitivity classes: very high, high, moderate, low, and very low. 10.65% of study area has very high sensitivity towards windstorm hazard which distributed at high area (Banjaran titiwangsa), West Coast area (Pahang and West Johor) and north area (Perlis). In addition, some part in the surrounding region of Peninsular Malaysia were also identified as windstorm hazard due to the dense of urban area such at Johor Bharu and Klang Valley area. 7.20% of the study area has very-low windstorm hazard and is mainly located in the middle part.

The new ACS membership model: a year in review

The Committee on Membership Affairs (MAC), with the support of the ACS Council and ACS staff, launched a new membership model in 2022. The model has different membership options to suit different needs. Members can now opt for a slimmed-down set of benefits with the Standard Package, which is half the price of the Premium Package. There is also a no-cost Basic Package for community associates. The cost of ACS membership also went down for all members. MAC remains confident in this new model and will continue to refine the full suite of available options and benefits according to our ongoing assessments of the ACS community’s needs. Overall membership in the society grew last year. We ended 2022 with over 173,000 individuals in our global membership community, compared with just over 151,000 a year earlier. But the inaugural year of our new membership model was not without challenges, which included

Comment: The new ACS membership model: a year in review

Comment: The new ACS membership model: a year in review

The Impact of Capability on Cooperative Performance and Sustainability: A Case Study of West Java Coffee Farmers Cooperatives

The industrial revolution in the agricultural sector is faced with demographic challenges (ageing), so intelligent technology, artificial intelligence, big data, and augmented reality are strategic choices. The application of information technology has a direct or indirect impact on Indonesia's agricultural sector, so coffee agribusiness cooperatives' information and communication technology (ICT) is also increasingly needed to respond to market demand. Along with the development of automation, physical, cyber systems and IoT, coffee cooperatives need to be ready and adapt to the era of the industrial revolution 4.0 in capturing market demand opportunities, value chains, and the added value of coffee agribusiness. ICT's presence can improve the performance and institutional sustainability of coffee farmers in rural areas, including cooperatives. Performance is a description of the achievement of organizational tasks to realize the goals, objectives, mission and vision of the organization and its business. The performance of the cooperative describes the services provided to its members and the community/farmers as measured by using indicators to see the working mechanism of the organization and its business that has been carried out and to determine the achievement of the objectives obtained. Meanwhile, the ability to innovate is essential, especially the use of ICT, which is deemed necessary to support the performance and sustainability of coffee cooperatives, so that cooperatives can compete, be stable, or survive in the era of increasing free competition. . Quantitative descriptive method with Linear Regression analysis was used to identify the effect of cooperative capability on the performance and sustainability of cooperatives in several coffee cooperatives in West Java. The study's results found that the power of the cooperative did not significantly affect the performance of the Coffee Cooperative. Still, the capability of the cooperative had a positive and significant effect on the sustainability of the Coffee Cooperative. At the same time, the performance of the cooperative has a positive and significant impact on the sustainability of the coffee cooperative. The ability of cooperatives has a positive and significant effect on sustainability mediated by the performance of the Coffee Cooperative in the Industrial Revolution 4.0 era. This study recommends increasing the capabilities of cooperatives in membership, management, and business models to become sustainable cooperatives based on information technology to achieve added value from market-driven coffee products.

Heterogeneity in mode choice behavior: A spatial latent class approach based on accessibility measures

We propose a method to estimate mode choice models, where preference parameters are sensitive to the spatial context of the trip origin, challenging traditional assumptions of spatial homogeneity in the relationship between travel modes and the built environment. The framework, called Spatial Latent Classes (SLC), is based on the integrated choice and latent class approach, although instead of defining classes for the decision maker, it estimates the probability of a location belonging to a class, as a function of spatial attributes. For each Spatial Latent Class, a different mode choice model is specified, and the resulting behavioral model for each location is a weighted average of all class-specific models, which is estimated to maximize the likelihood of reproducing observed travel behavior. We test our models with data from Portland, Oregon, specifying spatial class membership models as a function of local and regional accessibility measures. Results show the SLC increases model fit when compared with traditional methods and, more importantly, allows segmenting urban space into meaningful zones, where predominant travel behavior patterns can be easily identified. We believe this is a very intuitive way to spatially analyze travel behavior trends, allowing policymakers to identify target areas of the city and the accessibility levels required to attain desired modal splits.

ACR Presidential Address: Rheumatologists-Folks You Can Trust.

ACR Presidential Address: Rheumatologists-Folks You Can Trust.

The Effect of Performance on the Sustainability of Coffee Farmers’ Cooperatives in the Industrial Revolution 4.0 in West Java Indonesia

This research evaluates the effect the industrial revolution era 4.0 has had on the performance and sustainability of coffee farmer cooperatives, especially in the West Java region, considering that West Java is a coffee center area that is starting to be taken into account in Indonesia and the world in general, this is evidenced by the increasing number of farmer cooperatives coffee in the West Java region. One of the most monumental effects of the industrial revolution 1.0 was the birth of the British cooperative in Rochdale, the world’s first modern cooperative. The industry continues to develop, cooperatives continue to adapt and try to continue to develop in accordance with technological advances, especially now that the industrial revolution 4.0 has entered, which emphasizes information technology, especially the use of the internet. However, there has been no previous research examining the use of information technology in coffee farmer cooperatives, especially in the era of the industrial revolution 4.0, even though this information technology is very important in the development of cooperatives today. Quantitative descriptive method with SEM analysis was used to identify the effect of cooperative capability on the performance and sustainability of cooperatives in several coffee cooperatives in West Java. The study’s results found that the power of the cooperative did not significantly affect the performance of the Coffee Cooperative. Still, the capability of the cooperative had a positive and significant effect on the sustainability of the Coffee Cooperative. At the same time, the performance of the cooperative had a positive and significant impact on the sustainability of the coffee cooperative. The ability of cooperatives has a positive and significant effect on sustainability mediated by the performance of the Coffee Cooperative in the Industrial Revolution 4.0 era. This study recommends increasing the capabilities of cooperatives in membership, management, and business models to become sustainable cooperatives based on information technology to achieve added value from market-driven coffee products. The results of the study show that the respondents perceive that the cooperative’s performance has a positive effect on their sustainability. Improving the performance of cooperatives in terms of (1) productivity, (2) effectiveness, (3) quality, and (4) achievement can be regarded as having a direct impact on the Coffee Cooperative’s long-term viability. The industrial revolution in the agricultural sector is facing demographic challenges (ageing), so intelligent technology, artificial intelligence, big data, and augmented reality are strategic choices. The application of information technology has a direct or indirect positive impact on Indonesia’s agricultural sector. Cooperatives in coffee agribusiness is also increasingly needed to respond to market demand by applying Information and Communication Technology (ICT).

Modeling historic environmental pollutant exposures and non-Hodgkin lymphoma risk

Environmental exposures to chemicals are suspected risk factors for non-Hodgkin lymphoma (NHL), but few studies have assessed historic environmental risk factors. In this study, we estimated the associations between NHL and 1) historic environmental pollutant emissions from the Risk Screening Environmental Indicators (RSEI) model, which uses a database from the Environmental Protection Agency of toxic release emissions to air, water, and land, and 2) chemical mixtures measured in house dust (groups of PCBs, PAHs, and two mixtures of pesticides) for study participants enrolled in the NCI-SEER population-based case-control study (1998–2000) at four SEER centers – Detroit, Iowa, Los Angeles County, and Seattle. We assigned 11 years of annual temporally-varying historic environmental exposure scores by intersecting residential locations from participants' residential histories with a fine grid from the RSEI model and by performing inverse distance weighting between facilities releasing specific carcinogenic chemicals and residential locations for spatially-precise exposure assignments. We used Bayesian index low-rank kriging multiple membership models to identify important lag times for RSEI scores, cumulative effects of RSEI scores, and specific carcinogenic chemical releases into the environment. We found a significant positive association between RSEI scores and NHL at the maximum time lag of 11 years (OR = 1.17, 95% CI (1.06, 1.32)) and a significant cumulative RSEI score effect (OR = 1.30, 95% CI (1.02, 1.84)) for long-term residents in Detroit, where benzene and trichloroethylene were the most important chemicals driving this association. Additionally, we identified significant inverse associations for two study centers and time lags that did not persist in cumulative exposure models. Large weights for dichloromethane and pentachlorophenol in models of cumulative exposure also support evidence for their association with NHL risk. These results underscore the importance of considering historic and cumulative environmental exposures and using residential histories for diseases with long latency periods such as NHL.

Estimating Mixed Memberships in Directed Networks by Spectral Clustering

Community detection is an important and powerful way to understand the latent structure of complex networks in social network analysis. This paper considers the problem of estimating community memberships of nodes in a directed network, where a node may belong to multiple communities. For such a directed network, existing models either assume that each node belongs solely to one community or ignore variation in node degree. Here, a directed degree corrected mixed membership (DiDCMM) model is proposed by considering degree heterogeneity. An efficient spectral clustering algorithm with a theoretical guarantee of consistent estimation is designed to fit DiDCMM. We apply our algorithm to a small scale of computer-generated directed networks and several real-world directed networks.