Personalized travel sequence recommendation in urban environments poses a complex challenge due to dynamic spatial conditions, user diversity and limited resources. Traditional systems often overlook the multi-user dimension and fail to adapt to fluctuating crowd patterns. This work proposes a novel Multi-Agent Reinforcement Learning (MARL) framework that coordinates personalized itinerary planning while managing urban congestion. The framework comprises autonomous Travel Agents representing individual users and a centralized Congestion Management Authority that guides agents toward balanced spatial distributions. To enhance scalability and learning efficiency we introduce an interest-based clustering mechanism that groups users with similar preferences. Each cluster shares a policy network and experience buffer, and is trained using a Deep Q-Network (DQN) algorithm. This design significantly reduces memory consumption by 80% and training time by 67%. Compared to the baseline model our MARL framework improves interest alignment by 49.1% (0.85 vs. 0.57), PoI popularity by 50.9% (0.80 vs. 0.53) and reduces travel time by 50.5% (11.16% vs. 22.55%) resulting in more relevant, attractive and time-efficient itineraries. A real-world case study in Dubai validates the framework's ability to generate high-quality itineraries that align with user interests and mitigate overcrowding at popular sites. Additionally, a user study confirms improved satisfaction and perceived personalization compared to baseline methods. The results highlight the potential of MARL as a scalable, adaptive solution for next-generation spatial recommendation systems.

Social networks have been transforming the way people express opinions, post and react to events, and share ideas. Over the last decade, several studies on event detection from social media have been proposed, with the aim of extracting specific types of events, such as, social gatherings, natural disasters, and emergency situations, among others. However, these works do not consider the continuous processing of events over the social data streams, and therefore, cannot determine the spatial and temporal evolution of such events. This paper introduces a big data platform for event discovery, while tracking their evolution over space and time. We propose a scalable and efficient architecture that can manage and mine a huge data flow of unstructured streams, in order to detect geo-social events. The extracted clusters of events are indexed by a spatio-temporal index structure. We conduct experiments over twitter datasets to measure the effectiveness and efficiency of our system with respect to the existing major event detection techniques. An initial demonstration of our platform highlights its major advantage for detecting and tracking events spatially and temporally, thus allowing for great opportunities from application perspectives.

The tremendous rise of location-enriched microblogging has made it possible to discover social events from social media, as well as their evolution over space and time. Over the last decade, multiple studies on event detection from social media have been proposed, with the aim of extracting specific types of events, such as, social gatherings, natural disasters, and emergency situations, among others. However, existing works do not consider the incremental and continuous processing of events over the large amounts of social streams, and therefore, cannot determine the spatial and temporal evolution of such events. This work presents a big data mining platform for the incremental discovery of geo-social events based on a scalable and efficient architecture that can manage and mine a huge data flow of unstructured streams. We demonstrate our early results over twitter datasets and discuss its main advantage by incorporating advanced features for event extraction, thus allowing for great opportunities from application perspectives.

Nowadays, social network sites; such as Facebook and Twitter, have tremendous number of users in their repositories. Having this huge amount of data requires analyzing them to get statistics about the users and their interests. In this paper, we propose a new algorithm that clusters the nodes in social networks into communities based on their geodesic location and the similarity between their interests. The algorithm is examined thoroughly to test its performance. The experiments show that the algorithm achieves a high community detection accuracy.

K-nearest neighbors (KNN) search in a high-dimensional vector space is an important paradigm for a variety of applications. Despite the continuous efforts in the past years, algorithms to find the exact KNN answer set at high dimensions are outperformed by a linear scan method. In this paper, we propose a technique to find the exact KNN image objects to a given query object. First, the proposed technique clusters the images using a self-organizing map algorithm and then it projects the found clusters into points in a linear space based on the distances between each cluster and a selected reference point. These projected points are then organized in a simple, compact, and yet fast index structure called array-index. Unlike most indexes that support KNN search, the array-index requires a storage space that is linear in the number of projected points. The experiments show that the proposed technique is more efficient and robust to dimensionality as compared to other well-known techniques because of its simplicity and compactness. © 2011 Wiley Periodicals, Inc.