English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Nowadays, Online Travel Agencies provide the main service for bookingholidays, business trips, accommodations, etc. As in many e-commerce serviceswhere users, items, and preferences are involved, the use of a RecommenderSystem facilitates the navigation of the marketplaces. One of the mainchallenges when productizing machine learning models (and in this case,Learning-to-Rank models) is the need of, not only consistent pre-processingtransformations, but also input features maintaining a similar scale both attraining and prediction time. However, the features' scale does not necessarilystay the same in the real-world production environment, which could lead tounexpected ranking order. Normalization techniques such as featurestandardization, batch normalization and layer normalization are commonly usedto tackle the scaling issue. However, these techniques. To address this issue,in this paper we propose a novel scale-invariant ranking function (dubbed asSIR) which is accomplished by combining a deep and a wide neural network. Weincorporate SIR with five state-of-the-art Learning-to-Rank models and comparethe performance of the combined models with the classic algorithms on a largedata set containing 56 million booked searches from the Hotels.com website.Besides, we simulate four real-world scenarios where the features' scale at thetest set is inconsistent with that at the training set. The results reveal thatwhen the features' scale is inconsistent at prediction time, Learning-To-Rankmethods incorporating SIR outperform their original counterpart in allscenarios (with performance difference up to 14.7%), while when the features'scale at the training and test set are consistent our proposal achievescomparable accuracy to the classic algorithms.

A scale invariant ranking function for learning-to-rank: a real-world use case