Results

**1 - 3**of**3**### doi:10.1093/comjnl/bxv031

, 2014

"... In many applications, independence of event occurrences is assumed, even if there is evidence for dependence. Capturing dependence leads to complex models, and even if the complex models were superior, they fail to beat the simplicity and scalability of the independence assumption. Therefore, many m ..."

Abstract
- Add to MetaCart

(Show Context)
In many applications, independence of event occurrences is assumed, even if there is evidence for dependence. Capturing dependence leads to complex models, and even if the complex models were superior, they fail to beat the simplicity and scalability of the independence assumption. Therefore, many models assume independence and apply heuristics to improve results. Theoretical explanations of the heuristics are seldom given or generalizable. This paper reports that some of these heuristics can be explained as encoding dependence in an exponent based on the generalized harmonic sum. Unlike independence, where the probability of subsequent occurrences of an event is the product of the sin-gle event probability, harmony is based on a product with decaying exponent. For independence, the sequence probability is p1+1+···+1 = pn, whereas for harmony, it is p1+1/2+···+1/n. The generalized harmonic sum leads to a spectrum of harmony assumptions. This paper shows that harmony assump-tions naturally extend probability theory. An experimental evaluation for information retrieval (IR; term occurrences) and social networks (SN’s; user interactions) shows that assuming harmony is more suitable than assuming independence. The potential impact of harmony assumptions lies beyond IR and SN’s, since many applications rely on probability theory and apply heuristics to com-pensate the independence assumption. Given the concept of harmony assumptions, the dependence between multiple occurrences of an event can be reflected in an intuitive and effective way.

### MAP MAP MAP MAP

, 2012

"... What is our IR-driven mathematical framework (tool box) to investigate theoretically — to fully understand — why which model is better when? 3 / 37Intro & Background TF IDF IDF and LM P(t) P(q|d) and P(d|q): Conjunctive P(q|d) and P(d|q): Disjunctive Summary Appendix The world according to Binom ..."

Abstract
- Add to MetaCart

(Show Context)
What is our IR-driven mathematical framework (tool box) to investigate theoretically — to fully understand — why which model is better when? 3 / 37Intro & Background TF IDF IDF and LM P(t) P(q|d) and P(d|q): Conjunctive P(q|d) and P(d|q): Disjunctive Summary Appendix The world according to Binomial/Poisson Prob and Independent Events

### An Attribute-based Model for Semantic Retrieval

"... This paper introduces a knowledge-oriented approach for modelling semantic search. The modelling approach represents both semantic and textual data in one unifying framework, referred to as the probabilistic object-relational content modelling framework. The framework facilitates the transformation ..."

Abstract
- Add to MetaCart

This paper introduces a knowledge-oriented approach for modelling semantic search. The modelling approach represents both semantic and textual data in one unifying framework, referred to as the probabilistic object-relational content modelling framework. The framework facilitates the transformation of “term-only ” retrieval models into “semantic-aware ” retrieval models that consist of semantic propositions, such as relationships and classification of objects. To illustrate this facility, an attribute-based retrieval model, referred to as TF-IEF-AF-IDF, is instantiated using the modelling framework. The effectiveness of the developed retrieval model is demonstrated using the Internet Movie Database test collection. Overall, the probabilistic objectrelational content model can guide how semantic search and semantic data are modelled. 1