Rolling Up Random Variables in Data Cubes
Source
Evernote/IFTTT Feedly/Rolling Up Random Variables in Data Cubes.md
Summary
이 논문은 데이터 큐브(Data Cubes) 내의 이산 확률 변수에 계층적 구조(roll-ups)를 부여하는 새로운 기법을 제시합니다. Chow-Liu 알고리즘을 일반화하여, 큐브의 집계 구조를 통해 변수들의 결합 분포를 간결하게 근사하는 방식으로 roll-up을 구현합니다. 웹 트래픽 스트림의 이상 징후 모니터링 및 보고를 위한 실제 사례로 기법의 유효성을 입증합니다.
Key Points
- 데이터 큐브의 계층적 집계(roll-up) 구조를 확률 변수의 결합 분포 근사에 적용
- Chow-Liu 접근법의 일반화를 통한 간결한(parsimonious) 확률 모델링
- 웹 트래픽 이상 징후 감지 및 보고 시스템에 대한 실제 적용 사례 제시
Related
-
Similarity-based Clustering by Left-Stochastic Matrix Factorization
-
Neighborhood Preserving Codes for Assigning Point Labels: Applications to Stochastic Search
-
Fast Near-Duplicate Image Detection Using Uniform Randomized Trees
-
Behavior-Oriented Data Resource Management in Medical Sensing Systems
-
λ-Diverse Nearest Neighbors Browsing for Multidimensional Data
-
Protecting Sensitive Labels in Social Network Data Anonymization
-
Generalized Optimal Response Time Retrieval of Replicated Data from Storage Arrays
-
Point Representation for Local Optimization: Towards Multi-Dimensional Gray Codes
-
Information-Theoretic Outlier Detection for Large-Scale Categorical Data
-
An Unsupervised Feature Selection Framework for Social Media Data
-
Efficient Inference and Structured Learning for Semantic Role Labeling
-
Efficient Closed-Form Solution to Generalized Boundary Detection
-
Fast, Accurate Detection of 100,000 Object Classes on a Single Machine (Technical Supplement)
-
Continuous Birdsong Recognition Using Gaussian Mixture Modeling of Image Shape Features
-
Anomaly Extraction in Backbone Networks Using Association Rules
-
A Hamming Embedding Kernel with Informative Bag-of-Visual Words for Video Semantic Indexing
-
Regularized Latent Semantic Indexing (RLSI): 대규모 토픽 모델링을 위한 새로운 접근법
-
Efficient Estimation of Word Representations in Vector Space
-
Structured Streaming Skeleton (SSS): 온라인 인간 제스처 인식용 새로운 특징 추출 방법
-
Efficient Multiview Maintenance under Insertion in Huge Social Networks
-
Moment-Based Spectral Analysis of Large-Scale Networks Using Local Structural Information
-
Read-Performance Optimization for Deduplication-Based Storage Systems in the Cloud
-
Behavioural reconfigurable and adaptive data reduction in body sensor networks
-
Smooth Nonnegative Matrix Factorization for Unsupervised Audiovisual Document Structuring
-
Social Event Classification via Boosted Multimodal Supervised Latent Dirichlet Allocation
-
Supporting Flexible, Efficient, and User-Interpretable Retrieval of Similar Time Series
-
Online Graph Edge-Coloring in the Random-Order Arrival Model
-
Unsupervised Spatial Event Detection in Targeted Domains with Applications to Civil Unrest Modeling
-
Weakly Supervised Learning of Object Segmentations from Web-Scale Video
-
Knowledge-Based Approaches to Information Management in Coalition Environments
-
Joint consideration of energy-efficiency and coverage-preservation in microsensor networks
-
Active Learning through Adaptive Heterogeneous Ensembling (AHE)