* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download relevance feedback algorithms inspired by quantum detection
Relativistic quantum mechanics wikipedia , lookup
Delayed choice quantum eraser wikipedia , lookup
Coherent states wikipedia , lookup
Quantum decoherence wikipedia , lookup
Double-slit experiment wikipedia , lookup
Orchestrated objective reduction wikipedia , lookup
Path integral formulation wikipedia , lookup
Renormalization group wikipedia , lookup
Many-worlds interpretation wikipedia , lookup
Quantum entanglement wikipedia , lookup
Bell's theorem wikipedia , lookup
Theoretical and experimental justification for the Schrödinger equation wikipedia , lookup
Measurement in quantum mechanics wikipedia , lookup
Copenhagen interpretation wikipedia , lookup
Quantum computing wikipedia , lookup
History of quantum field theory wikipedia , lookup
EPR paradox wikipedia , lookup
Density matrix wikipedia , lookup
Quantum teleportation wikipedia , lookup
Interpretations of quantum mechanics wikipedia , lookup
Quantum group wikipedia , lookup
Canonical quantization wikipedia , lookup
Quantum machine learning wikipedia , lookup
Quantum key distribution wikipedia , lookup
Hidden variable theory wikipedia , lookup
Symmetry in quantum mechanics wikipedia , lookup
Bra–ket notation wikipedia , lookup
Quantum cognition wikipedia , lookup
Quantum state wikipedia , lookup
RELEVANCE FEEDBACK ALGORITHMS INSPIRED BY QUANTUM DETECTION Abstract Information Retrieval (IR) is concerned with indexing and retrieving documents including information relevant to a user’s information need. Relevance Feedback (RF) is a class of effective algorithms for improving Information Retrieval (IR) and it consists of gathering further data representing the user’s information need and automatically creating a new query. In this paper, we propose a class of RF algorithms inspired by quantum detection to re-weight the query terms and to rerank the document retrieved by an IR system. These algorithms project the query vector on a subspace spanned by the eigenvector which maximizes the distance between the distribution of quantum probability of relevance and the distribution of quantum probability of non-relevance. The experiments showed that the RF algorithms inspired by quantum detection can outperform the state-of-the-art algorithms. Existing System The automatic procedure that modify the user’s queries is known as RF; some relevance assessments about the retrieved documents are collected and the query is expanded by the terms found in the relevant documents, reduced by the terms found in the irrelevant documents or reweighted using relevant or irrelevant documents. RF can be positive, negative or both. Positive RF only brings relevant documents into play and negative RF makes only use of irrelevant documents; any effective RF algorithms includes a “positive” component. Although positive feedback is a well established technique by now, negative feedback is still problematic and requires further investigation, yet some proposals have already been made such as grouping irrelevant documents before using them for reducing the query Proposed System It is designed to compute the new query vector using a linear combination of the original vectors, the relevant document vectors and the non-relevant document vectors, where the labels of relevance are collected in a training set. Quantum probability is the theory of probability developed within Quantum Mechanics (QM). In QM, a probability space can be represented as vectors, matrices and operators between them. A tutorial would be out of the scope of this paper, therefore we provide the information instrumental to understanding the rest of this paper. Detection consists of identifying the information concealed in the data which are transmitted by the source placed on one side, through a channel to the detector placed on the other side. The data are only a representation of the “true” information that one side wants to transmit. Implementation Implementation is the stage of the project when the theoretical design is turned out into a working system. Thus it can be considered to be the most critical stage in achieving a successful new system and in giving the user, confidence that the new system will work and be effective. The implementation stage involves careful planning, investigation of the existing system and it’s constraints on implementation, designing of methods to achieve changeover and evaluation of changeover methods. Modules 1.Vector Space Model 2.Relevance Feedback 3.Quantum Probability 4.Quantum Detection Vector Space Model The VSM for IR represents both documents and queries as vectors of the kdimensional real space Rk. This vector space is defined by k basis vectors corresponding to the terms extracted from a document collection. Each document vector results from the weighted linear combination of the basis vectors which represents the terms extracted from the document collection. The early formulation of the VSM was reported by Salton who later developed the model in the 1970s for describing the statistical methods to measure semantic relationships between words such as synonymy and polysemy and to build networks of terms and documents. The VSM was revisited and was then experimented and applied to several tasks (e.g., crosslanguage IR , passage retrieval and automatic hypertext generation. Relevance Feedback The RF algorithm is also known as Rocchio’s algorithm and it is designed to compute the new query vector using a linear combination of the original vectors, the relevant document vectors and the non-relevant document vectors, where the labels of relevance are collected in a training set. Relevance feedback is a feature of some information retrieval systems. The idea behind relevance feedback is to take the results that are initially returned from a given query and to use information about whether or not those results are relevant to perform a new query. Quantum Probability A probability space is given by some observables and by a probability function of these observables. Quantum probability is the theory of probability developed within Quantum Mechanics (QM). In QM, a probability space can be represented as vectors, matrices and operators between them. When a probability function is provided, each observable value and then each basis vector corresponds to a probability measure given by the function, thus obtaining a probability distribution. The Gleason theorem explains why the trace rule is used to compute the probabilities of events represented by vector spaces. The theorem basically states that the density matrix introduced in this section can encapsulate all the information about a probability space, that is, it provides a probability distribution for any conceivable observable. Quantum Detection Detection consists of identifying the information concealed in the data which are transmitted by the source placed on one side, through a channel to the detector placed on the other side. The data are only a representation of the “true” information that one side wants to transmit. When the particle arrives at the other side, it is measured by the receiver. This measurement is accomplished by an observable. The observed values are utilized to determine the state of the signal (e.g., a document) given by the coder. In IR, the information that is relevant to the user’s information need is transmitted by a system to the user by means of a document which is only a representation of the information fulfilling the user’s need. The values observed from a document (e.g., term frequency) can serve to decide about the state of the document. Although the use of PRF with the algorithms inspired by quantum detection can be less effective than RF and less effective than the algorithms with no expansion for larger n’s for long or verbose queries. When inspecting the results at the level of query size, the situation appears to be more complex than explicit RF because the algorithms that are inspired by quantum detection were more effective than the baseline algorithms depending on n or query size. System Configuration H/W System Configuration: - Pentium –IV RAM - 1GB Hard Disk - 80 GB Processor S/W System Configuration: Operating System :Windows95/98/2000/XP Application Server : Tomcat5.0/6.X Front End : HTML, Java, Jsp Scripts Server side Script : Java Server Pages. Database : Mysql Database Connectivity : JDBC. : JavaScript.