Identifying labels for image collections

A system comprising:

• a data processing apparatus;
• and
• a data store storing instructions that, when executed by the data processing apparatus, cause the data processing apparatus to perform operations comprising:
• obtaining a collection of images;
• obtaining, for each image in the collection of images, image similarity data that indicates a measure of similarity of content depicted in the image to content depicted in other images in the collection of images;
• generating, based on the similarity data, two or more image clusters from the collection of images, each image cluster including one or more images from the collection of images;
• for each image cluster:
• obtaining, for each image in the image cluster, a set of image labels;
• generating, from each set of image labels obtained for each image in the image cluster, a set of cluster labels;
• selecting one or more cluster labels from the set of cluster labels;
• identifying the selected cluster labels as a set of collection labels for the collection of images;
• identifying one or more suggested image collections based on the set of collection labels, each of the suggested image collections being different from the collection of images, wherein the identifying the one or more suggested image collections includes,
• obtaining, for each of a plurality of candidate image collections, a corresponding candidate set of collection labels;
• obtaining, for each candidate set of collection labels, set similarity data that indicates a measure of similarity of the candidate set of collection labels to the set of collection labels;
• selecting, based on the set similarity data of the candidate sets of collection labels, one or more candidate image collections;
• and
• identifying the selected candidate image collections as suggested image collections;
• receiving from a user device, a selection of one or more suggested image collections to include with the collection of images;
• and
• providing, to the user device, the selected one or more suggested image collections to be displayed along with the collection of images.

A non-transitory computer readable medium comprising instructions that, when executed by a data processing apparatus, cause the data processing apparatus to perform operations comprising:

• obtaining a collection of images;
• obtaining, for each image in the collection of images, image similarity data that indicates a measure of similarity of content depicted in the image to content depicted in other images in the collection of images;
• generating, based on the similarity data, two or more image clusters from the collection of images, each image cluster including one or more images from the collection of images;
• for each image cluster:
• obtaining, for each image in the image cluster, a set of image labels;
• generating, from each set of image labels obtained for each image in the image cluster, a set of cluster labels;
• selecting one or more cluster labels from the set of cluster labels;
• identifying the selected cluster labels as a set of collection labels for the collection of images;
• identifying one or more suggested image collections based on the set of collection labels, each of the suggested image collections being different from the collection of images, wherein the identifying the one or more suggested image collections includes,
• obtaining, for each of a plurality of candidate image collections, a corresponding candidate set of collection labels;
• obtaining, for each candidate set of collection labels, set similarity data that indicates a measure of similarity of the candidate set of collection labels to the set of collection labels;
• selecting, based on the set similarity data of the candidate sets of collection labels, one or more candidate image collections;
• and
• identifying the selected candidate image collections as suggested image collections;
• receiving, from a user device, a selection of one or more suggested image collections to include with the collection of images;
• and
• providing, to the user device, the selected one or more suggested image collections to be displayed along with the collection of images.