Web and Data Science

Get Started. It's Free
or sign up with your email address
Web and Data Science by Mind Map: Web and Data Science

1. Onto

1.1. [VL1/VL2 Logic, Logic, Logic]

1.1.1. FOL

1.1.1.1. Semantics

1.1.1.1.1. Interpretation

1.1.1.1.2. Entailment

1.1.1.2. Syntax

1.1.1.2.1. Terms

1.1.1.2.2. Formulars

1.1.1.3. Compactness of Topology

1.1.1.4. Löwenheim-Skolem-Property

1.1.1.5. First Lindström Theorem

1.1.1.6. Pro und Contra von FOL

1.1.1.7. Complexity

1.1.1.7.1. FOL in PSPACE

1.1.1.7.2. FOL in AC^0 (pratical Scenarios)

1.1.2. Proof Calculi

1.1.2.1. Resolution

1.1.2.1.1. Prenex Normal Form

1.1.2.1.2. Skolem Form (Skolemization)

1.1.2.1.3. Clausal Normal Form

1.1.2.2. Correctness & Completness

1.1.2.3. Semi-Decidability

1.2. [VL3/VL4 Finite Model Theory]

1.2.1. Inexpressibility Tools

1.2.1.1. Games

1.2.1.1.1. Ehrenfeucht-Fraissé Games

1.2.1.1.2. Quantifier Rank

1.2.1.2. Locality

1.2.1.2.1. Bounded Number of Degree Property

1.2.1.2.2. Gaifman Locality

1.2.1.2.3. Hanf Locality

1.2.1.3. Reductions

1.2.1.4. FOL has the 0-1 Law

1.2.2. Fixed-Point Logics

1.2.2.1. Datalog

1.3. [VL5/VL6 Data Exchange]

1.3.1. Main Setting

1.3.1.1. Consistency

1.3.1.2. Materialization

1.3.1.3. Goodness

1.3.2. Materialization

1.3.2.1. Marked NULLs

1.3.3. Representations

1.3.4. Certain Answears

1.3.5. Solutions

1.3.5.1. Universal Solutions

1.3.5.1.1. Homomorphism

1.3.5.2. Core Solutions

1.3.6. Mappings

1.3.6.1. Source Schema

1.3.6.2. Target Schema

1.3.6.3. Source-target dependencies

1.3.6.3.1. st-tgd

1.3.6.4. Target dependencies

1.3.6.4.1. tgd

1.3.6.4.2. edg

1.3.7. Chase

1.3.7.1. Terminating

1.3.7.1.1. With Solution

1.3.7.1.2. Without Solution

1.3.7.2. Not Terminating

1.3.7.2.1. Simple Dependency Graphs

1.4. [VL7/VL8 Ontology Based Data Access]

1.4.1. Closed Word Assumption

1.4.1.1. NULLs

1.4.2. Open World Assumption

1.4.3. Ontologies

1.4.3.1. Signature

1.4.3.1.1. Constants

1.4.3.1.2. Concepts

1.4.3.1.3. Roles

1.4.3.2. Terminological Box (TBox)

1.4.3.3. Assertional Box (ABox)

1.4.4. Reasoning Services

1.4.4.1. Model

1.4.4.2. Satisfiability

1.4.4.3. Coherence

1.4.4.4. Concept satisfiability

1.4.4.5. Subsumption

1.4.4.6. Instance Check

1.4.4.7. Query Answering

1.4.5. Certain Answers

1.4.6. Descriptions Logics (DLs)

1.4.6.1. Lightweight DLs

1.4.7. Tableaux

1.4.8. Rewriting

1.4.8.1. Perfect Rewriting

1.4.8.1.1. Termination

1.4.8.1.2. Correctnis

1.4.8.1.3. Completness

1.4.8.2. Unifications

1.4.8.3. Annonymization

1.4.8.4. Reduction

1.4.9. Unfolding

1.5. [VL9/VL10 Ontology Integration]

1.5.1. Motivation

1.5.1.1. Import

1.5.1.2. Merge

1.5.1.3. Versioning

1.5.2. Belief Revision

1.5.2.1. Consequence Operator

1.5.2.1.1. Tarskian Consequence Operator

1.5.2.2. Belief Set

1.5.2.3. AGM Belief Revision

1.5.2.3.1. Expansion

1.5.2.3.2. Contraction

1.5.2.3.3. Revision

1.5.2.3.4. Criticism

1.5.2.4. Remainder Set

1.5.2.5. Selection Function

1.5.2.5.1. Partial Meet

1.5.3. Ontology Change

1.5.3.1. Goals

1.5.3.1.1. Overcome Heterogenity

1.5.3.1.2. Combine Ontologies

1.5.3.1.3. Modify Ontologies

1.5.3.2. Operations

1.5.3.2.1. Ontology Mapping

1.5.3.2.2. Ontology Merge

1.5.3.2.3. Ontology Evolution

1.5.3.2.4. Ontology Learning

1.5.4. Iterated Belief Revision

1.6. [VL11 Stream Processing]

2. Data Mining

2.1. [VL1-3]: Bayesian Networks

2.1.1. Agent

2.1.1.1. Rational

2.1.1.2. Autonomous

2.1.1.3. Social

2.1.1.4. Simple information retrieval

2.1.2. Triple Tradeoff

2.1.2.1. Complexity/Capacity of H: c(H)

2.1.2.2. Training set size: N

2.1.2.3. Generalization error: E

2.1.2.4. As N increases, E decreases

2.1.2.5. As c(H) increases, first E decreases, then E increases again (overfitting)

2.1.3. Conditional Probability

2.1.4. Bayes Rule

2.1.5. Bayesian Network

2.1.5.1. Conditional Probability Table (CPT)

2.1.5.2. Full Joint Distribution

2.1.5.3. P(X_1, ..., X_n)

2.1.6. Inference Tasks

2.1.6.1. Simple queries

2.1.6.2. Conjunctive queries

2.1.6.3. Optimal decisions

2.1.6.4. Value of information

2.1.6.5. Sensitivity analysis

2.1.6.6. Explanation

2.1.7. Exact Inference

2.1.7.1. Inference by Enumeration

2.1.7.2. Variable Elimination

2.1.7.3. Pearl's Belief Propagation Algorithm

2.1.7.3.1. \lambda messages

2.1.7.3.2. \pi message

2.1.8. Approximate Inference

2.1.8.1. Sampling from empty network

2.1.8.2. Rejection sampling

2.1.8.3. Likelihood weighting

2.1.8.4. Markov Chain Monte Carlo

2.1.8.4.1. Markov Blanket

2.1.9. Learning Bayesian Networks

2.1.9.1. Known structure

2.1.9.1.1. Full Bayesian Learning

2.1.9.1.2. MAP Learning

2.1.9.1.3. Maximum Likelihood Learning

2.1.9.2. Unknown structure

2.1.9.2.1. Model selection

2.1.10. Learning BNs with hidden variables

2.1.10.1. Known structure

2.1.10.1.1. Expectation-Maximization

2.1.10.2. Unknown structure

2.1.10.2.1. Structural EM

2.2. [VL4-7]: Structured Causal Models

2.2.1. General concepts

2.2.1.1. Confounder

2.2.1.2. Mediator

2.2.1.3. Exogenous/Endogenous

2.2.1.3.1. Exogenous

2.2.1.3.2. Endogenous

2.2.1.4. Simpson's paradox

2.2.1.5. d-separation

2.2.1.6. Network structure

2.2.1.6.1. Chain

2.2.1.6.2. Fork

2.2.1.6.3. Collider

2.2.1.7. Equivalent graphs

2.2.1.7.1. V-Structure

2.2.1.7.2. Skeleton

2.2.1.8. Conditioning

2.2.2. IC algorithm

2.2.3. Intervention

2.2.3.1. Blocking conditions

2.2.3.2. Backdoor criterion

2.2.3.3. Frontdoor criterion

2.2.3.4. General Causal Effect

2.2.3.4.1. Inverse Probability Weighing

2.2.3.4.2. Average Causal Effect/Causal Effect Difference

2.2.3.5. Specific Causal Effect

2.2.3.6. Controlled Direct Effect

2.2.3.7. Adjusting

2.2.4. Linear SCMs

2.2.4.1. Exogenous variables for Linear SCMs

2.2.4.2. Total effect in Linear SCMs

2.2.4.3. Path coefficients

2.2.4.4. Regression coefficients

2.2.4.5. Instrumental Variable

2.2.4.5.1. Conditional Instrumental Variables

2.2.4.6. Instrumental Sets

2.2.4.7. Wright's Rule

2.2.5. Counterfactuals

2.2.5.1. Consistency rule

2.2.5.2. Counterfactuals can simulate intervention

2.2.5.3. Additive Intervention

2.3. [VL8]: Junction Trees

2.3.1. Junction Tree Property

2.4. [VL9]: Inference in Probabilistic Graphical Models

2.4.1. Markov Networks (=Markov Random Fields)

2.5. [VL10]: Computational Learning Theory

2.5.1. PAC Learning

2.5.1.1. Computational Complexity

2.5.1.2. Sample complexity

2.5.1.3. VC Dimension

2.5.1.4. Shattering

2.5.1.4.1. Dichotomy

2.5.2. Mistake Bounds

2.5.2.1. Optimal Mistake Bounds

2.6. [VL11]: Learning FOL definable concepts

2.6.1. Parameter Learning

2.6.2. Model Learning

3. Autonomous Agents and Information Retrieval

3.1. [VL1]: Agents and Documents

3.1.1. Documents

3.1.1.1. Document Zones

3.1.1.2. Incidence matrices for overlap matching

3.1.1.3. Term weighting

3.1.1.3.1. Term frequency

3.1.1.3.2. Term scarcity

3.1.1.3.3. Document length

3.1.1.4. Inverse document frequency

3.1.1.5. Vector space model

3.1.1.5.1. tf.idf

3.1.1.5.2. scarcity considered

3.1.1.5.3. document frequency considered

3.1.1.5.4. normalize to document length

3.1.1.6. Latent Semantic Indexing (LSI)

3.1.1.6.1. Dimension reduction

3.1.1.6.2. Sigular Value Decomposition (Eigenwertzerlegung)

3.1.1.7. Relevance feedback

3.1.1.7.1. Precision

3.1.1.7.2. Recall

3.1.1.7.3. F-Score

3.1.1.7.4. Kappa measure

3.1.1.7.5. Rocchio algorithm

3.2. [VL2]: Probabilistic Information Retrieval

3.2.1. Odds

3.2.2. Probability Ranking Principle

3.2.2.1. Binary Independence Retrieval

3.2.2.1.1. Retrieval Status Value (RSV)

3.2.2.2. Binary Independence Indexing

3.2.3. Bayesian Network Information Retrieval

3.2.4. Language Models

3.2.4.1. Ponte and Croft LM

3.2.4.2. Lavrenko and Croft LM

3.3. [VL3]: Topic Modeling

3.3.1. Naive Bayes: Plate Notation

3.3.1.1. C: topics/classes

3.3.1.2. N: number of words in considered documents

3.3.1.3. W_i: one specific word in corpus

3.3.1.4. M: documents

3.3.1.5. W: words now in document

3.3.2. Multinomial Naive Bayes

3.3.3. Probabilistic LSI/LSA

3.3.3.1. Hyperlink modeling using pLSI

3.3.4. Latent Dirichlet Allocation (LDA)

3.3.4.1. Dirichlet Distributions

3.3.4.2. Smoothed LDA

3.3.4.3. Parameter Learning

3.3.4.3.1. Variatonal EM

3.3.4.3.2. Gibbs Sampling

3.3.4.4. Hyperlink modeling using LDA

3.3.5. Relational Topic Model (RTM)

3.3.5.1. Collapsed Gibbs Sampling for RTM

3.4. [VL4]: Probabilistic Reasoning over Sequential Structures

3.4.1. Probablisitic Temporal Models

3.4.1.1. Dynamic Bayesian Networks

3.4.1.1.1. Parameter Learning via EM

3.4.1.2. Hidden Markov Models

3.4.1.3. Kalman Filters

3.4.2. Inference Tasks

3.4.2.1. Filtering

3.4.2.2. Prediction

3.4.2.3. Smoothing

3.4.2.4. Most likely explanation

3.4.3. Viterbi Algorithm

3.5. [VL5]: Word Semantics and Latent Relational Structures

3.5.1. Pointwise Mutual Information

3.5.1.1. (P)PMI-Matrix

3.5.2. Word representation: word2vec

3.5.2.1. Continuous Bag of Words (CBOW)

3.5.2.2. Skip-Grams (SG)

3.5.2.2.1. Leaning method: Negative Sampling (SGNS)

3.5.2.3. GloVe

3.5.2.4. Dynamic Context Windows

3.5.2.5. Context Distribution Smoothing

3.6. [VL6]: Multi-Relational LSA

3.6.1. Polarity Inducing LSA (PILSA)

3.6.2. Multi-Relational LSA (MRLSA)

3.6.2.1. Encode Relations in Tensor

3.6.2.2. Tensor decomposition (sim. to SVD)

3.7. [VL7]: Probablisitic Relational Models

3.7.1. Markov Networks (MRFs)

3.7.2. Markov Logic Networks (MLNs)

3.7.3. Inductive Logic Programming

3.7.3.1. Inductive Learning: FOIL

3.8. [VL8]: Lifted Inference in PRMs

3.8.1. Lifted Junction Tree Algorithm

3.9. [VL9]: Game Theory / Social Choice Theory

3.9.1. Game Theory

3.9.1.1. Game

3.9.1.1.1. Agents

3.9.1.1.2. Actions

3.9.1.1.3. Outcomes

3.9.1.1.4. Payoffs

3.9.1.2. Multiagent Systems: Criteria

3.9.1.2.1. Social welfare

3.9.1.2.2. Surplus

3.9.1.2.3. Pareto efficiency

3.9.1.2.4. Individual rationality

3.9.1.2.5. Stability

3.9.1.2.6. Symmetry

3.9.1.3. Strategies

3.9.1.3.1. Dominant strategies

3.9.1.4. Vickrey Auctions

3.9.1.5. Nash Equilibrium

3.9.2. Social Choice Theory

3.9.2.1. Agenda paradox

3.9.2.2. Condorcet principle

3.9.2.3. Borda count principle

3.9.2.3.1. Inverted-order paradox

3.9.2.4. Arrow's Theorem

3.10. [VL10]: Mechanism Design

3.10.1. Direct mechanisms

3.10.2. Incentive-compatible

3.10.3. Dominant strategies implementation

3.10.3.1. Revelation principle

3.10.4. Social choice functions

3.10.5. Groves Mechanism