Report on the TREC-3 Experiment: A Learning Scheme in a Vector Space Model
نویسندگان
چکیده
This paper describes and evaluates a retrieval scheme, or more precisely an additional retrieval mechanism based on interdocument relationships, that can be integrated in almost all existing retrieval schemes (e.g., Boolean, hybrid Boolean, vector-processing or probabilistic models). The intent of our approach consists of inferring knowledge about document contents based on the relevance assessments of past queries. Through a learning process, our scheme establishes relevance links between documents found relevant for the same request. Based on this information and a list of retrieved records for the current request, the proposed mechanism tries to improve the ranking of the retrieved items in a sequence most likely to satisfy user intent. The underlying hypothesis of this mechanism states that future requests addressed to the system should have some degree of similarity with previous queries, or that the retrieval apparatus will process requests for which it has already found a partial, appropriate answer in the past.
منابع مشابه
A New Formulation for Cost-Sensitive Two Group Support Vector Machine with Multiple Error Rate
Support vector machine (SVM) is a popular classification technique which classifies data using a max-margin separator hyperplane. The normal vector and bias of the mentioned hyperplane is determined by solving a quadratic model implies that SVM training confronts by an optimization problem. Among of the extensions of SVM, cost-sensitive scheme refers to a model with multiple costs which conside...
متن کاملSpace Vector Control Scheme of Three Level ZSI Applied to Wind Energy Systems
In this paper the Space Vector Control Scheme is implemented for a Wind Energy System using Three Level Impedance Source Inverter (ZSI). The wind energy system uses a Self Excited Induction generator (SEIG) which is the most emerging application in the field of Wind Energy Conversion System (WECS). The proposed system is modelled with a generator-side Diode Bridge Rectifier and a Stand-Alone si...
متن کاملApplying SIGMA to the TREC-7 Filtering Track
This paper presents the research method and results from applying SIGMA (System of Information Gathering Market-based Agents) to the adaptive filtering task of the TREC-7 Filtering Track. Our work in SIGMA is based on the research hypothesis that a multi-agent learning approach, where each agent learns a local model for information filtering, performs better than a single-agent learning approac...
متن کاملReport on the TREC-4 Experiment: Combining Probabilistic and Vector-Space Schemes
This paper describes and evaluates a retrieval scheme combining the OKAPI probabilistic retrieval model with various vector-space schemes. In this study, each retrieval strategy represents both queries and documents using the same set of single terms; however they weight them differently. To combine these search schemes, we do not apply a given combination operator on the retrieval status nor t...
متن کاملForecasting the Tehran Stock market by Machine Learning Methods using a New Loss Function
Stock market forecasting has attracted so many researchers and investors that many studies have been done in this field. These studies have led to the development of many predictive methods, the most widely used of which are machine learning-based methods. In machine learning-based methods, loss function has a key role in determining the model weights. In this study a new loss function is ...
متن کامل