Item Recommendations Task
(Redirected from Item Recommendation Task)Jump to navigation Jump to search
- Input: an Item Set, such as a product catalog.
- output: a Ranked Item List (of rank values for each ranked item).
- measures: such as Clickthrough Rates, and NDGC.
- It can be solved by an Item Recommendations System that applies an (item recommendations algorithm).
- It can be supported by an Item Relevance Prediction Task.
- It can range from being a Heuristic Item Recommendation Task to being a Data-Driven Item Recommendation Task.
- It can range from being a Anchor-based Recommendation Task (such as a complementary recommendations) to being a Anchor-free Recommendation Task.
- It can range from being a Personalized Recommendation Task to being a Non-Personalized Recommendation Task, if user data is provided for the user.
- It can range from being a Contextual Recommendation Task to being a Non-Contextual Recommendation Task, if contextual data is provided for the recommendation context (e.g. time-of-day).
- It can range from being a Single-Item Recommendation Task to being a Multi-Item Recommendation Task.
- It can range from being an Engagement-Maximizing Recommendation Task to being a Profit-Maximizing Recommendation Task.
- It can be instantiated in an Item Recommendation Act.
- It can range from being a Collaborative Filtering-based Item Recommendation Task (if users-item interaction data is provided) to being an Item Metadata-based Recommendation Task (if item metadata is provided).
- a Product Recommendation Task, such as:
- Social Recommending, such as people recommending/recommending friends and community recommending/recommending groups;
- Job Recommending;
- Text Item Recommending, such as book recommending webpage recommending, and news recommending;
- a Contextual Advertising Task;
- an Item(s) Recommendation Benchmark Task, such as a Netflix Prize Task.
- See: Candidate Set Generation, Query Suggestion, Advice, Consumer Engagement Measure, Item Relevance Scoring.
- (Melville & Sindhwani, 2011) ⇒ Prem Melville, and Vikas Sindhwani. (2011). “Recommender Systems.” In: (Sammut & Webb, 2011) p.829
- QUOTE: The goal of a recommender system is to generate meaningful recommendations to a collection of users for items or products that might interest them. Suggestions for books on Amazon, or movies on Netflix, are real-world examples of the operation of industry-strength recommender systems. The design of such recommendation engines depends on the domain and the particular characteristics of the data available. For example, movie watchers on Netflix frequently provide ratings on a scale of 1 (disliked) to 5 (liked). Such a data source records the quality of interactions between users and items. Additionally, the system may have access to user-specific and item-specific profile attributes such as demographics and product descriptions, respectively. Recommender systems differ in the way they analyze these data sources to develop notions of affinity between users and items, which can be used to identify well-matched pairs. Collaborative Filtering systems analyze historical in …
- (Bogers, 2009) ⇒ Toine Bogers. (2009). “Recommender Systems for Social Bookmarking." PhD Thesis, Tilburg University.
- QUOTE: … More specifically, we want to investigate the task of item recommendation, where interesting and relevant items — bookmarks or scientific articles — are retrieved and recommended to the user, based on a variety of information sources about the user and the items. … The main contribution of this thesis is a principled investigation of the usefulness of different algorithms and information sources for recommending relevant items to users of social bookmarking services. … We show how prevalent these phenomena are, propose methods for automatically detecting them, and examine the influence they might have on the item recommendation task.
- (Kitts et al., 2000) ⇒ Brendan Kitts, David Freed, and Martin Vrieze. (2000). “Cross-Sell: A fast promotion-tunable customer-item recommendation method based on conditionally independent probabilities.” In: Proceedings of the sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2000). doi:10.1145/347090.347181