Managing and sharing item metadata For a high level overview of the item terminology and item metadata data schema, please refer to the Item Concepts page. Now that we’ve introduced the high level concept of items in the Froomle platform, we will give a more in depth and advanced view of how Froomle ingests, transforms, stores and uses your item metadata. Preparing item metadata Before you send your metadata to Froomle we ask you to format it according to one of our Item Metadata Templates so we can properly ingest it. If necessary we are flexible to perform translations from your structure to our standard, provided you clearly describe which mapping needs to be done. For more information about these different templates please refer to the Item Concepts page. Sharing item metadata We offer multiple options for sharing metadata as described on the Integration Handbook main page. What metadata does Froomle need? Your item metadata is an ever-changing collection where old items disappear, new items are added and existing items are updated. By using the previously described methods and formats of metadata delivery Froomle can keep track of all these changes and keep an up to date view of your own item catalog and its metadata. Froomle uses this view of your item metadata for various purposes: Determining which items are available at the current time Depending on your use case we can use availability of an item’s metadata in the latest dump (Retail) or attributes like publication time and publication end time (News & Social Network) to determine the availability of an item. We then use this availability information to control if an item should or should not be recommended at the current time. Applying filters to recommendation requests Using the Configurations API (which can be managed by you or by Froomle) constraints can be configured and saved in advance to apply them during a recommendation request in the future. A multitude of configurations can be stored at the same time allowing you to dynamically change the look and feel of your requested recommendations depending on the context of the recommendations. Examples of such filters are: Only recommend items within the same category as the current item. Do not recommend products with a price lower than €20. Only recommend products that are currently discounted. Do not recommend news articles with a publication time older than 24 hours. Content based recommendations Contrary to collaborative filtering (use browsing behavior of other users for recommendations), any content based recommendation techniques need metadata to work. If you want to test these algorithms or use specific modules that require these algorithms, we need metadata. This can be labels/categories of any items but also mood, complexity, length of articles. Providing metadata in our generated recommendations When you request recommendations we will always answer with an item identifier, score and rank. By supplying metadata we can include any of the additionally supplied fields in our responses. This can likely save you a database round-trip and make your website or app more responsive for your users. Improving analysis and reporting Numbers only tell us so much. Interesting relations and correlations can be uncovered more easily when there is metadata that you can link to each item. We extensively use metadata like title, tags, categories, … during our own analysis to gain new insights and plan our next move in your personalization story. When reporting back to you we incorporate as much information as possible to provide clear reporting that also explains why and isn’t just limited to a single improved metric. Personalizing your search experience Next to regular item recommendations we also offer a personalized search solution which combines our industry leading recommendation engine with the power of Elasticsearch. Compared to regular recommendations, metadata is a fundamental requirement for building a search engine.