Supervised learning methods for diversification of image search results

Date

2020-01-01

Author

Goynuk, Burak
Altıngövde, İsmail Sengör

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

126
views

0
downloads

© Springer Nature Switzerland AG 2020.We adopt a supervised learning framework, namely R-LTR [17], to diversify image search results, and extend it in various ways. Our experiments show that the adopted and proposed variants are superior to two well-known baselines, with relative gains up to 11.4%.

URI

https://hdl.handle.net/11511/56712

DOI

https://doi.org/10.1007/978-3-030-45442-5_20

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Explicit diversification of search results across multiple dimensions for educational search Yigit-Sert, Sevgi; Altıngövde, İsmail Sengör; Macdonald, Craig; Ounis, Iadh; ULUSOY, ÖZGÜR (Wiley, 2020-09-01) Making use of search systems to foster learning is an emerging research trend known assearch as learning. Earlier works identified result diversification as a useful technique to support learning-oriented search, since diversification ensures a comprehensive coverage of various aspects of the queried topic in the result list. Inspired by this finding, first we define a new research problem, multidimensional result diversification, in the context of educational search. We argue that in a search engine for th...
Parallelization of Robust Multigrid Technique Using OpenMP Technology Martynenko, Sergey; Zhou, Weixing; Gökalp, İskender; Bakhtin, Vladimir; Toktaliev, Pavel (2021-01-01) © 2021, Springer Nature Switzerland AG.This article represents the parallel multigrid component analysis of Robust Multigrid Technique (RMT). The RMT has been developed for black-box solution of a large class of (non)linear boundary value problems in computational continuum mechanics. Parallel RMT can be constructed by combination of the algebraic and geometric approaches to parallelization. The geometric smoother-independent approach based on a decomposition of the given problem into 3κ (κ= 1, 2, … ) subpr...
Camera trapping large mammals in yenice forest habitats : a feasibility study for camera trapping large mammals in Yenice Forests, Turkey Can, Özgün Emre; Togan, İnci Zehra; Department of Biology (2008) Widely applicable, quantitative field methods are needed to gather wildlife data for conservation and management initiatives in Turkey. In order to evaluate the use of camera traps in forest habitats of Turkey, we conducted a 5 phase camera trap survey by using 16 passive infrared-triggered cameras with a total sampling effort of 1200 camera trap days in Yaylacık Research Forest, a 50 km2 forest patch of Yenice Forest in Karabük during January-May 2006. The camera trap survey confirmed the presence of grey ...
The role of gaze as a deictic cue in human robot interaction Yılmaz, Efecan; Fal, Mehmetcan; Acartürk, Cengiz (2020-01-01) © Springer Nature Switzerland AG 2020.Gaze has a major role in social interaction. As a deictic reference, gaze aims at attracting visual attention of a communication partner to a referred entity in the environment. Gaze direction in natural faces is a well-investigated domain of research at behavioral and neurophysiological levels. However, our knowledge about deictic role of gaze in Human Robot Interaction is limited. The present study focuses on a comparative analysis of the deictic role of gaze directio...
Goal Orientation in Human Computer Interaction Tasks: An Experimental Investigation of User Engagement and Interaction Modalities Fal, Mehmetcan; Acartürk, Cengiz (2021-01-01) © 2021, Springer Nature Switzerland AG.Developing natural and intuitive Human Computer Interaction is a primary target of interaction design. The analysis of a user’s engagement to the task is an essential component for the assessment of the interaction for design efficiency. The investigation of task engagement requires considering users’ goal orientation since performance-oriented and learning-oriented individuals exhibit motivational differences. Learning-oriented individuals focus on progress and master...

Citation Formats

B. Goynuk and İ. S. Altıngövde, “Supervised learning methods for diversification of image search results,” 2020, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/56712.