데이터셋 상세
미국
Discriminative Mixed-Membership Models
Although mixed-membership models have achieved great success in unsupervised learning, they have not been widely applied to classification problems. In this paper, we propose a family of discriminative mixed-membership models for classification by combining unsupervised mixed membership models with multi-class logistic regression. In particular, we propose two variants respectively applicable to text classification based on latent Dirichlet allocation and usual feature vector classification based on mixed membership naive Bayes models. The proposed models allow the number of components in the mixed membership to be different from the number of classes. We propose two variational inference based algorithms for learning the models, including a fast variational inference which is substantially more efficient than mean-field variational approximation. Through extensive experiments on UCI and text classification benchmark datasets, we show that the models are competitive with the state of the art, and can discover components not explicitly captured by the class labels.
데이터 정보
연관 데이터
경기도 데이터분석 분석 데이터셋 분류 내역
공공데이터포털
이 데이터는 경기도 데이터분석시스템에서 제공하는 분석 데이터셋 분류 내역 자료로, 데이터셋의 관리 및 분류 체계를 체계적으로 기록한 메타데이터 성격의 정보입니다. 주요 항목에는 데이터셋ID, 데이터셋분류코드, 대표분류여부, 등록자ID, 등록일시, 수정자ID, 수정일시 등이 포함되어 있으며, 각 데이터셋이 어떠한 주제 영역이나 코드 체계로 분류되는지를 확인할 수 있습니다. 이 데이터는 행정, 정책, 경제, 사회 등 다양한 분석 분야에서 데이터셋의 활용 범위를 파악하는 데 기초 자료로 쓰일 수 있습니다.
SA1-based B27 Family Blending as at 2011-08-11
공공데이터포털
SA1 based data for Family Blending, for 2011 Census. Count of couple families with children on Census night based on place of usual residence. Excludes 'Couple families with no children', 'One parent families' and 'Other families'. Excludes families in: 'Non-family/Non-classifiable households', 'Non-private dwellings' and 'Migratory, off-shore and shipping' SA1s. Data sourced from: http://www.abs.gov.au/census. For further information about these and related statistics, contact the National Information and Referral Services on 1300 135 070. Periodicity: 5-Yearly.
경기도 데이터분석 분석 데이터셋 분류 유형 코드
공공데이터포털
이 데이터는 경기도 데이터분석시스템에서 사용되는 분석 데이터셋 분류 유형 코드를 정의한 메타데이터입니다. 주요 항목으로는 번호, 데이터셋분류유형코드, 데이터셋분류유형명, 등록자ID, 등록일시, 수정자ID, 수정일시가 포함됩니다. 이 데이터는 데이터셋의 분류 체계를 식별하고, 코드 기반으로 다른 데이터셋과 연계할 수 있도록 지원합니다. 활용 방안으로는 행정기관의 데이터 관리 체계화, 연구기관의 주제별 데이터 접근 용이화, 민간의 데이터 서비스 개발을 위한 코드 매핑 등이 있습니다.
School Bus Report October 2021 21-993
공공데이터포털
For the 2020-2021 school year, only students who were enrolled in blended learning were assigned to bus routes. Therefore, students who are routed are assumed to be enrolled in blended learning. Students are reported as those who were planned and routed, but may or may not have used the service provided. Routes were determined by those routed with a vendor and student(s) assigned Vehicles are identified by school bus company reported data. This data is subject to data entry error and is dependent on the school bus companies to maintain. All transportation sites were included, and reported by type. A transportation site was defined by a unique location identified by a street address. Please note: multiple students or schools can be located in the same address and would therefore represent a stop for multiple students at the same site. MetroCard passes are assigned at the school. For public and charter schools, OPT reports students assigned to MetroCard, which includes those assigned a MetroCard by serial number or "T'd" for MetroCard, meaning the school indicated the student as eligible for a MetroCard but did not yet assign a serial number. For Non-Public schools, OPT reports on students who are eligible for MetoCards if they are not assigned to yellow bus service Some special education schools do not classify students by grade level. Students attending such schools are identified as “NG.” Service type is representative only of the type of busing a student is assigned to and is not reflective of the educational classification of a student. Students who are classified as special education but do not have an IEP mandate for specific transportation accommodations may be assigned to general education busing. General education students under certain circumstances including, but not limited to temporary housing status, orders of protection and medical conditions may be assigned to special education busing. OPT assigns students to busing but cannot confirm student ridership. This reporting would have to be captured at the school level. Students in temporary housing (STH) situations other than those reported to be living in a DHS shelter* were excluded from the report due to data quality issues. While the DOE student data system (ATS) has a housing indicator flag, it is often inaccurate and unreliable. Therefore, OPT cannot reliably report on the temporary housing status of students other than those residing in DHS shelters, where a daily report indicating shelter status is provided. *Transportation data for students in Foster Care is also provided. Please see section 21-993 B7 for reporting. Stop to school bus service is typically not provided to students after the 6th grade. Those assigned to this service are students who are granted individual or school-wide exceptions. In addition to students who applied for transportation because of foster care placement, also reported were students known to be in foster care who received busing or A MetroCard to provide a more complete report of students in foster care who received service. Students in foster care were identified by a monthly data feed from ACS to OPT via DIIT. Source is subject to error due to data matching to identify OSIS based on name and data of birth.