In this Competetion, Kagglers will develop models capable of classifying visual styles of graphic design images. A Style can be interpreted as a specific combination of relationships between elements in a graphic design work, Designers have mastered the skill to create visual experiences through composing a complex interplay between the content and style. Therefore, one high level motivation is to automate design style analysis to accelerate the understanding between AI and Design industry. Another motivation is to help enterprises obtain AI abilities to insight into the creative content and analyse the style factor affecting the conversion rate based on data return of different groups. \n\nExcept for traditional objects classification of an image, thus far the algorithmic basis of the process that predict visual style of images is unknown and there exists no artificial system with similar capabilities. One of the biggest challenges is that there is no universally accepted or standard definition of design styles. Because this is such a creative and interesting task, To this end, we put together a dataset that contains “representative” design style images so that you can build your skills with the common machine learning task. \n\nThe dataset for this competition consists of 3,000 images in total with style annotation. Images are collected from third-party image URls. We tagged them with style categories, which were carefully chosen considering different factors such as design technique, level of image emotion, average number of style instance. \nYou can browse all style tags images here.\nDisclaimer：The actual images are not owned by us. For researchers and educators who wish to use the images for non-commercial research and/or educational purposes, we just provide access through this competition site. \n\nIn this competition, you are challenged to build a model that is capable to predict styles of images,each image in the dataset is annotated with several style labels. We know that design styles are subjective, each style label needs to be voted by different viewers to gain consensus, we got some help from designers in differernt fields to test the style annotations of dataset, so the labels can be regarded as truevalue. You will use the dataset to train a classification model.
This is the home page of the competition. You don't need a subtitle here. The competition sub-title will appear above. This is where you introduce the problem. You can upload images using the \"select files\" widget on the left in the competition wizard. Upload an image, refresh the page, copy its URL, then insert within the wizard's editor.
\nIf you are copy-pasting from another application, like Word or your browser, try to make sure the html formatting is clean. You can view a page's html using the button at the top right of the editor's toolbar.
\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "The evaluation pages describes how submissions will be scored and how students should format their submissions. You don't need a sub-title at the top (the page title appears above). For an example of a high quality evaluation page, see the 2013 KDD Cup page, which we have copied here:
\r\nThe evaluation metric for this competition is Mean F1-Score. The F1 score, commonly used in information retrieval, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (tp) to all predicted positives (tp + fp). Recall is the ratio of true positives to all actual positives (tp + fn). The F1 score is given by:
\r\n\\[ F1 = 2\\frac{p \\cdot r}{p+r}\\ \\ \\mathrm{where}\\ \\ p = \\frac{tp}{tp+fp},\\ \\ r = \\frac{tp}{tp+fn} \\]
\r\nThe F1 metric weights recall and precision equally, and a good retrieval algorithm will maximize both precision and recall simultaneously. Thus, moderately good performance on both will be favored over extremely good performance on one and poor performance on the other.
\r\nFor every author in the dataset, submission files should contain two columns: AuthorId and DuplicateAuthorIds. DuplicateAuthorIds should be a space-delimited list. Every AuthorId counts as his/her own duplicate, and every duplicate should be listed under each of its respective ids. For example, if you suspect author A, B, and C are the same, you should list (A,A B C), (B,B A C), (C,C A B).
\r\nThe file should contain a header and have the following format:
\r\nAuthorId,DuplicateAuthorIds", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }
1,1
8,8
9,9 10
10,10 9
etc.