Summary of the paper
| Title | Model Summaries for Location-related Images |
|---|---|
| Authors | Ahmet Aker and Robert Gaizauskas |
| Abstract | At present there is no publicly available data set to evaluate the performanceof different summarization systems on the task of generating location-relatedextended image captions. In this paper we describe a corpus of human generatedmodel captions in English and German. We have collected 932 model summaries inEnglish from existing image descriptions and machine translated these summariesinto German. We also performed post-editing on the translated German summariesto ensure high quality. Both English and German summaries areevaluated using a readability assessment as in DUC and TAC to assess theirquality. Our model summaries performed similar to the ones reported in Dang(2005) and thus are suitable for evaluating automatic summarization systems onthe task of generating image descriptions for location related images. Inaddition, we also investigated whether post-editing of machine-translatedmodel summaries is necessary for automated ROUGE evaluations. We found a highcorrelation in ROUGE scores between post-edited and non-post-edited modelsummaries which indicates that the expensive process of post-editing is notnecessary. |
| Language | Machine Translation, SpeechToSpeech Translation |
| Topics | Corpus (creation, annotation, etc.), Summarisation, Machine Translation, SpeechToSpeech Translation |
Full paper  |
Model Summaries for Location-related Images |
| Bibtex | @InProceedings{AKER10.102, author = {Ahmet Aker and Robert Gaizauskas}, title = {Model Summaries for Location-related Images}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |