Encoding local correspondence in topic models
|Encoding local correspondence in topic models|
|Author(s)||Mehdi R.E., Mohamed Q., Mustapha A.|
|Published in||Proceedings - International Conference on Tools with Artificial Intelligence, ICTAI|
|Keyword(s)||Automatic Image Annotation, Local Influence, Probabilistic Graphical Models, Topic Models (Extra: Automatic image annotation, Label correlations, Latent variable modeling, Local influence, Multi-label learning, Probabilistic graphical models, Topic model, Wikipedia, Artificial intelligence, Image analysis, Tools, Image retrieval)|
|Article||BASE, CiteSeerX, Google Scholar|
|Web||Ask, Bing, Google (PDF), Yahoo!|
|Download and mirrors|
|Local copy||Not available|
|Remote mirror(s)||Not available|
|Export and share|
|BibTeX, CSV, RDF, JSON|
|Browse properties · List of conference papers|
Encoding local correspondence in topic models is a 2013 conference paper written in English by Mehdi R.E., Mohamed Q., Mustapha A. and published in Proceedings - International Conference on Tools with Artificial Intelligence, ICTAI.
Exploiting label correlations is a challenging and crucial problem especially in multi-label learning context. Labels correlations are not necessarily shared by all instances and have generally a local definition. This paper introduces LOC-LDA, which is a latent variable model that adresses the problem of modeling annotated data by locally exploiting correlations between annotations. In particular, we represent explicitly local dependencies to define the correspondence between specific objects, i.e. regions of images and their annotations. We conducted experiments on a collection of pictures provided by the Wikipedia 'Picture of the day' website, and evaluated our model on the task of 'automatic image annotation'. The results validate the effectiveness of our approach.
- This section requires expansion. Please, help!
Probably, this publication is cited by others, but there are no articles available for them in WikiPapers.