Popularity, novelty and relevance in point of interest recommendation: an experimental analysis

Abstract

Recommender Systems (RSs) are often assessed in off-line settings by measuring the system precision in predicting the observed user’s ratings or choices. But, when a precise RS is on-line, the generated recommendations can be perceived as marginally useful because lacking novelty. The underlying problem is that it is hard to build an RS that can correctly generalise, from the analysis of user’s observed behaviour, and can identify the essential characteristics of novel and yet relevant recommendations. In this paper we address the above mentioned issue by considering four RSs that try to excel on different target criteria: precision, relevance and novelty. Two state of the art RSs called and follow a classical Nearest Neighbour approach, while the other two, and are based on Inverse Reinforcement Learning. and optimise precision, tries to identify the characteristics of POIs that make them relevant, and , a novel RS here introduced, is similar to but it also tries to recommend popular POIs. In an off-line experiment we discover that the recommendations produced by and optimise precision essentially by recommending quite popular POIs. can be tuned to achieve a desired level of precision at the cost of losing part of the best capability of to generate novel and yet relevant recommendations. In the on-line study we discover that the recommendations of and are liked more than those produced by . The rationale of that was found in the large percentage of novel recommendations produced by , which are difficult to appreciate. However, excels in recommending items that are both novel and liked by the users.

Recency-based spatio-temporal similarity exploration for POI recommendation in location-based social networks

“…It is critical to understand the inherent preference pattern in the personalized recommendation, in addition to the trade-off existing between accuracy and fairness. Massimo and Ricci (2021) analyzed two aspects of this problem. First, the impact of biasing toward popular items on precise recommendations.…”

Estimation of Mass and Lengths of Sintered Workpieces Using Machine Learning Models

“…When these models are exposed to new (unseen) input data, they are required to generalize well: pretrained models (learned on previous computations on the input data) are expected to provide reliable and repeatable decisions on unseen data. Automating decisionmaking by using ML has already gained significant attention in recent years and use cases in banking (classification of the benign and fraudulent transaction [5]), healthcare (accurately monitoring the health of the patient [6]), and travel and transportation (hotel/visit recommender systems [21]) domains have been developed, just to name a few.…”

Identification of Urban Functional Areas and Their Mixing Degree Using Point of Interest Analyses

“…In contrast, POI data have the advantages of simple acquisition and convenient processing, meaning that POI data are widely used in geographical research. At present, the research on POI mainly focuses on a recommendation model [19][20][21] and urban functional zoning [22], and the research on POI itself mainly focuses on automatic classification [23] and classification accuracy by using machine learning. Nak introduced POI into eye-tracking data analysis [24], making the research scope of POI broader.…”