1.INTRODUCTION
Information models are significant because they are representative of three different mathematical models, with their own methods for representing documents and calculating similarity between documents and users’ profiles [4]. It will focus on three models: the Boolean Model, the Vector Space Model, and the Probabilistic Model. The Boolean Model is an instance of the set-theoretic models, where documents are represented as sets of words, on which operations are performed in order to determine similarities. The Vector Space Model is an algebraic model in which documents and users’ profiles are represented as vectors. Operations, such as the dot product of two vectors, are used to determine similarities as a scalar values. Finally, in the Probabilistic Model probabilistic inference is used to retrieve documents. This model relies on probabilistic theorems, such asBayes’ theorem, to compute similarities as probabilities of relevance. Three models supporting information retrieval were covered, with a particular emphasis on their mode of representation of the documents and their processing algorithms.
1.INTRODUCTIONInformation models are significant because they are representative of three different mathematical models, with their own methods for representing documents and calculating similarity between documents and users’ profiles [4]. It will focus on three models: the Boolean Model, the Vector Space Model, and the Probabilistic Model. The Boolean Model is an instance of the set-theoretic models, where documents are represented as sets of words, on which operations are performed in order to determine similarities. The Vector Space Model is an algebraic model in which documents and users’ profiles are represented as vectors. Operations, such as the dot product of two vectors, are used to determine similarities as a scalar values. Finally, in the Probabilistic Model probabilistic inference is used to retrieve documents. This model relies on probabilistic theorems, such asBayes’ theorem, to compute similarities as probabilities of relevance. Three models supporting information retrieval were covered, with a particular emphasis on their mode of representation of the documents and their processing algorithms.
การแปล กรุณารอสักครู่..
