7 Ranking
Ranking is a regression machine learning technique.
7.1 About Ranking
Ranking is a machine learning technique to rank items.
Ranking is useful for many applications in information retrieval such as e-commerce, social networks, recommendation systems, and so on. For example, a user searches for an article or an item to buy online. To build a recommendation system, it becomes important that similar articles or items of relevance appear to the user such that the user clicks or purchases the item. A simple regression model can predict the probability of a user to click an article or buy an item. However, it is more practical to use ranking technique and be able to order or rank the articles or items to maximize the chances of getting a click or purchase. The prioritization of the articles or the items influence the decision of the users.
The ranking technique directly ranks items by training a model to predict the ranking of one item over another item. In the training model, it is possible to have items, ranking one over the other by having a "score" for each item. Higher ranked items have higher scores and lower ranked items have lower scores. Using these scores, a model is built to predict which item ranks higher than the other.
7.2 Ranking Methods
Oracle Machine Learningsupports pairwise and listwise ranking methods through XGBoost.
For a training data set, in a number of sets, each set consists of objects and labels representing their ranking. A ranking function is constructed by minimizing a certain loss function on the training data. Using test data, the ranking function is applied to get a ranked list of objects. Ranking is enabled for XGBoost using the regression function. OML4SQL supports pairwise and listwise ranking methods through XGBoost.
Pairwise ranking: This approach regards a pair of objects as the learning instance.
The pairs and lists are defined by supplying the same
case_id value. Given a pair of
objects, this approach gives an optimal ordering for that pair.
Pairwise losses are defined by the order of the two objects. In
OML4SQL, the algorithm
uses LambdaMART to perform pairwise ranking with the goal of
minimizing the average number of inversions in ranking.
Listwise ranking: This approach takes multiple lists of ranked objects as learning instance. The items in a list must have the same case_id. The algorithm uses LambdaMART to perform list-wise ranking.
See Also:
- "Ranking Measures and Loss Functions in Learning to Rank" a research paper presentation at https://www.researchgate.net/
- Oracle Database PL/SQL Packages and Types Reference for a listing and explanation of the available model settings for XGBoost.
Note:
The term hyperparameter is also interchangeably used for model setting.Related Topics