In clique tree clustering, inference consists of propagation in a clique tree compiled from a Bayesian network. In this paper, we develop an analytical approach to characterizing clique tree growth as a function of increasing Bayesian network connectedness, specifically: (i) the expected number of moral edges in their moral graphs or (ii) the ratio of the number of non-root nodes to the number of root nodes. In experiments, we systematically increase the connectivity of bipartite Bayesian networks, and find that clique tree size growth is well-approximated by Gompertz growth curves. This research improves the understanding of the scaling behavior of clique tree clustering, provides a foundation for benchmarking and developing improved BN inference algorithms, and presents an aid for analytical trade-off studies of tree clustering using growth curves. Reference: O. J. Mengshoel, "Macroscopic Models of Clique Tree Growth for Bayesian Networks." In Proc. of the 22nd National Conference on Artificial Intelligence (AAAI-07). July 2007, Vancouver, Canada, pp. 1256-1262. BibTex Reference: @inproceedings{mengshoel07macroscopic, author = "Mengshoel, O. J.", title = "Macroscopic Models of Clique Tree Growth for {Bayesian} Networks", year = "2007", booktitle = {Proceedings of the Twenty-Second National Conference on Artificial Intelligence (AAAI-07)}, pages = "1256-1262", address = "Vancouver, British Columbia" }

The networks are already pruned using the largest K–core value, and are obtained from the Network Repository [37].

This bundle contains:

• instance collections we used for the experimental evaluation in FlatZinc format (as well as the MiniZinc models we used for running CP-SAT),
• and raw logs of all experimental runs reported in the paper.

This table contains 5 series, with data for years 1871 - 1971 (not all combinations necessarily have data for all years). This table contains data described by the following dimensions (Not all combinations are available): Unit of measure (1 items: Persons ...) Geography (1 items: Canada ...) Population (5 items: Total population; Population in incorporated centres of 30,000 to 99,999 persons; Population in incorporated centres of 5,000 to 29,999 persons; Population in incorporated centres of 100,000 persons and over ...).

TO DOWNLOAD THE DATASET, CLICK ON THE "Download" BUTTON

In a series of seven studies, we examined how participants interpreted and responded to "intermediate" social groups. Specifically, the studies captured reactions to novel groups designed to manipulate social group intermediacy (Studies 1-5), Black/White biracial people (Study 6), and bisexual people (Study 7). In each study, participants provided separate evaluations of an intermediate group and two comparison groups (e.g., Black/White biracial people, Black people, White people). Intermediate groups were consistently rated as less conceptually legitimate (e.g., less distinctive, not a "real" group) than other groups.These studies are described together in a single research report (Burke, Perry, Dovidio, & LaFrance). For details about the procedures, please consult that report and its accompanying supplementary document. For convenience, a draft (preprint) of the manuscript (84 pages) and a draft of the supplementary document (22 pages) are included with this deposit as a single PDF file (106 pages total). More recent versions of these documents may be available elsewhere.All of the experiments were conducted online and all employed convenience samples. Most participants were recruited via Amazon Mechanical Turk. Study 3 (the one that manipulated the population size of the intermediate group) and Study 4 (the one that manipulated how the spots were inherited) used Yale School of Management's eLab service, which permits people to sign up to participate in surveys in exchange for fixed probabilities of earning gift cards.The dates of the individual studies are listed below. The numbering (e.g., Study 1, Study 2) corresponds to the order of presentation in the research report, which deviates from the order in which the studies were conducted.Pilot - manipulation of color: February 2014Pilot - manipulation of spot size: April 2014Study 1 - novel groups control: July 2014 (manipulated intermediate/control and choice/trait)Study 2 - novel groups medium spots: May 2014 (manipulated presentation of intermediate group - big & small spots versus medium spots and early versus late in the procedure)Studies 3 and 4 - novel groups group size and inherit: January 2015 (manipulated group size and how the spots were inherited)Study 5 - novel groups ingroup: May 2015 (manipulated whether participant was an outside observer or a member of one group)Study 6 - racial groups: August 2014 (White, Black, biracial, Native American)Study 7 - sexual orientation groups: October 2014 (heterosexual, gay/lesbian, bisexual, asexual)These data files contain missing values coded as -99 in addition to the usual missing values that are simply empty. A value of -99 indicates that the participant saw the item but did not respond to it.I recognize that the documentation of this project could be more detailed. I have many projects that could benefit from additional public documentation. If you are using the data, please reach out to me via email and I can prioritize sharing more information about this one. My most up-to-date email address should always be available at saraemilyburke.com/contact.html.

These data were generated as part of a two-and-a-half-year ESRC-funded research project examining the digitalisation of higher education (HE) and the educational technology (Edtech) industry in HE. Building on a theoretical lens of assetisation, it focused on forms of value in the sector, and governance challenges of digital data. It followed three groups of actors: UK universities, Edtech companies, and investors in Edtech. The researchers first sought to develop an overview of the Edtech industry in HE by building three databases on Edtech companies, investors in Edtech, and investment deals, using data downloaded from Crunchbase, a proprietary platform. Due to Crunchbase’s Terms of Service, only parts of one database are allowed to be submitted to this repository, i.e. a list of companies with the project’s classification. A report offering descriptive analysis of all three databases was produced and is submitted as well. A qualitative discursive analysis was conducted by analysing seven documents in depth. In the second phase, researchers conducted interviews with participants representing three groups of actors (n=43) and collected documents on their organisations. Moreover, a list of documents collected from Big Tech (Microsoft, Amazon, and Salesforce) were collected to contextualise the role of global digital infrastructure in HE. Due to commercial sensitivity, only lists of documents collected about investors and Big Tech are submitted to the repository. Researchers then conducted focus groups (n=6) with representatives of universities (n=19). The dataset includes transcripts of focus groups and outputs of writing by participants during the focus group. Finally, a public consultation was held via a survey, and 15 participants offered qualitative answers.

Data on broad age groups and gender for the population 15 years of age and older in private households of Canada, provinces and territories.

As of January 2025, 24.2 percent of Facebook users in the United States were aged between 25 and 34 years, making up Facebook’s largest audience in the country. Overall, 19 percent of users belonged to the 18 to 24-year age group. Does everyone in the U.S. use Facebook? In 2023, there were approximately 247 million Facebook users in the U.S., a figure which is projected to steadily increase, and reach 262.8 million by 2028. Social media users in the United States have a very high awareness of the social media giant. Expectedly, 94 percent of users had heard of the brand in 2023. Although the vast majority of U.S. social networkers knew of Facebook, the likeability of the platform was not so impressive at 68 percent. Nonetheless, usage, loyalty, and buzz around the brand remained relatively high. Facebook, Meta, and the metaverse A strategic rebranding from Facebook to Meta Platforms in late 2021 boded well for the company in Mark Zuckerberg’s attempt to be strongly linked to the metaverse, and to be considered more than just a social media company. According to a survey conducted in the U.S. in early 2022, Meta Platforms is the brand that Americans most associated with the metaverse.  

As of February 2025, 37.5 percent of X’s (formerly Twitter) global audience was aged between 25 and 34 years. The second-largest age group demographic on the platform was represented by users aged between 18 and 24 years, with a share of 32.1 percent. Users aged less than 18 years accounted for two percent of users, while those aged 50 or older accounted for roughly 7.3 percent. X is a male-dominated platform As of January 2024, more than 60 percent of X users were male. Although all mainstream social media platforms tend to have a slightly more male-skewing audience, X stands out above Instagram, Snapchat, TikTok, and Facebook when it comes to user gender demographics. Overall, Pinterest is the only mainstream platform to have a higher share of female users. X Blue for you It is not uncommon for social media users to now have the chance to become subscribers of their chosen online networks for a monthly fee. X Blue is a subscription service from X that gives users special benefits and features. A blue verification mark, edit post functionality, fewer ads, priority ranking in chats, and longer video upload times are some of the perks offered.

The once-a-decade decennial census was conducted in April 2010 by the U.S. Census Bureau. This count of every resident in the United States was mandated by Article I, Section 2 of the Constitution and all households in the U.S. and individuals living in group quarters were required by law to respond to the 2010 Census questionnaire. The data collected by the decennial census determine the number of seats each state has in the U.S. House of Representatives and is also used to distribute billions in federal funds to local communities. The questionnaire consisted of a limited number of questions but allowed for the collection of information on the number of people in the household and their relationship to the householder, an individual's age, sex, race and Hispanic ethnicity, the number of housing units and whether those units are owner- or renter-occupied, or vacant. Results for sub-state geographic areas in New Mexico were released in a series of data products. These data come from Summary File 1 (SF-1). The geographic coverage for SF-1 includes the state, counties, places (both incorporated and unincorporated communities), tribal lands, school districts, census tracts, block groups and blocks, among others. The data in these particular RGIS Clearinghouse tables are for Quay County and all census block groups within the county. There are two data tables in this file. Table DC10_00628 shows the number of households by the following categories--total households; total family households; husband-wife family households; total other family households; male householder, no wife present; female householder, no husband present; total nonfamily households; nonfamily households with householder living along; and finally, nonfamily households with householder not living alone. Table DC10_00629 shows percent distribution of households for each of these same categories. These files, along with file-specific descriptions (in Word and text formats) are available in a single zip file.

Continuous Register Statistics: Population by communities, age (five-year groups), Spanish, Sex and Year. Annual. Autonomous Communities and Cities.

Estonia Number of Enterprise Groups data was reported at 6,988.000 Unit in 2017. This records an increase from the previous number of 6,660.000 Unit for 2016. Estonia Number of Enterprise Groups data is updated yearly, averaging 4,912.500 Unit from Dec 2004 (Median) to 2017, with 14 observations. The data reached an all-time high of 6,988.000 Unit in 2017 and a record low of 3,221.000 Unit in 2004. Estonia Number of Enterprise Groups data remains active status in CEIC and is reported by Statistics Estonia. The data is categorized under Global Database’s Estonia – Table EE.O012: Number of Enterprise Groups: by Owner Country.

This 2011 Population Census dataset contains statistics relevant to demographic, household, educational, economic and housing characteristics of the Hong Kong population residing in the 1620 Large Street Block Groups in 2011. The dataset also contains the boundaries of individual Large Street Block Groups. Since 1961, a population census has been conducted in Hong Kong every 10 years and a by-census in the middle of the intercensal period. The 2011 Population Census, which was conducted in June to August 2011, provides benchmark statistics on the socio-economic characteristics of the Hong Kong population vital to the planning and policy formulation of the government. This dataset will be incorporated into Population Distribution Framework Spatial Data Theme.

This table contains the data of the cliques (complete synteny subgraphs) of BAHD and SCP/SCPL genes

Continuous Register Statistics: 07com003.px. National. Population by sex, municipalities and age (five year groups).

Data on employment income statistics by industry groups (4-digit code) from the North American Industry Classification System (NAICS) 2017, visible minority, highest level of education, work activity during the reference year, age and gender, for the population aged 15 years and over who reported weeks worked and employment income in 2020 in private households in Canada, provinces and territories.

Recall, precision and F-measure are defined as follows: Recall (R) , Precision (P) and F-measure , where TP, FP, and FN represent true positive, false positive, and false negative, respectively.

Self-reported social mixing patterns are commonly used in mathematical models of infectious diseases. It is particularly important to quantify patterns for school-age children given their disproportionate role in transmission, but it remains unclear how the structure of such social interactions changes over time. By integrating data collection into a public engagement programme, we examined self-reported contact networks in year 7 groups in four UK secondary schools. We collected data from 460 unique participants across four rounds of data collection conducted between January and June 2015, with 7,315 identifiable contacts reported in total. Although individual-level contacts varied over the study period, we were able to obtain out-of-sample accuracies of more than 90% and F-scores of 0.49–0.84 when predicting the presence or absence of social contacts between specific individuals across rounds of data collection. Network properties such as clustering and number of communities were broadly consistent within schools between survey rounds, but varied significantly between schools. Networks were assortative according to gender, and to a lesser extent school class, with the estimated clustering coefficient larger among males in all surveyed co-educational schools. Our results demonstrate that it is feasible to collect longitudinal self-reported social contact data from school children and that key properties of these data are consistent between rounds of data collection.