Query expansion is a common technique used to improve retrieval effectiveness. In this paper, we propose a novel query expansion technique based on concept clique for Markov network information retrieval model. This technique strengthens the simple relationships between terms in the following two ways:(1) terms in a clique express a similar concept and will be added into query expansion, so that it is effective to expanded into some terms with low similar to query terms but highly related to query topic; (2) query term dependencies are used to select concept cliques as candidates. The selection of concept cliques based on a connected graph is effective to avoid topic drift during expanding polysemous query terms. Experiments on several collections show that new approach makes significant improvements and more effective on collections with polysemous terms.
JPEG2000 is the image compression standard which includes three core techniques: wavelet lifting scheme and EBCOT and MQ arithmetic coder. The EBCOT algorithm uses a Wavelet transform to generate the subband samples, which are quantized and coded. Normal Wavelet transforms fail to represent such functions effectively while Ridge functions are effective in representing functions that have discontinuities along straight lines. If the coefficients obtained from Ridgelet transform of the image with more discontinuities along straight lines have to be subject to quantization and coding process with EBCOT, the existing structure of the EBCOT should be modified to suit with the output of the Finite Ridgelet Transform (FRIT). In this paper, an effective image coding technique is proposed. The technique, which involves transforming the image into another domain with Ridgelet function and then quantizing and coding the coefficients with modified EBCOT, is similar to JPEG2000. A modified EBCOT algorithm for FRIT coefficients has been also proposed in this paper. Experimental results show the combination of modified EBCOT with FRIT which is used to code novel images such as images with straight lines and galaxy images significantly outperforms the combination of EBCOT with Wavelet Transform in novel image coding in terms of PSNR.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.