Improving Itinerary Recommendations for Tourists Through Metaheuristic Algorithms: An Optimization Proposal

Technological developments in recent years led to the emergence of increasingly sophisticated recommender systems to support multi-day travel itineraries that fall under the Tourist Trip Design Problem (TTDP). Various problem analogies are widely used to solve TTDP, such as Traveling Salesman Problem (TSP), Vehicle Routing Problem (VRP), Orienteering Problem (OP), and Team Orienteering Problem with Time Windows (TOPTW). For multi-day route recommendation, TOPTW is suitable as a problem analogy since there is a per-day travel duration constraint. So far, TTDP with TOPTW does not consider the weighting (priority level of users) for each requirement attribute in a multi-attribute-based TOPTW to ensure personalized recommendations. In addition, running time remains a challenge in many studies in the TOPTW area. Many metaheuristic algorithms have been adopted to TOPTW for generating a time-efficient approach. Komodo Mlipir Algorithm (KMA) emerges as a new algorithm that promises good scalability. Therefore, we propose KomoTrip, a method that adopts the discrete version of KMA and Multi-Attribute Utility Theory (MAUT) to recommend optimal travel routes per day by accommodating the multi-attribute preferences of users. We perform three evaluation scenarios, i.e., general performance, Degree of Interest (DOI) combinations, and varying numbers of Points of Interest (POI), consistently demonstrating that KomoTrip outperforms several benchmark algorithms in terms of computational time efficiency and also exhibits robust fitness values across different problem dimension scales. Thus, KomoTrip can be regarded as an efficient algorithm to recommend optimal multi-day tour routes, effectively incorporating weighted multi-attribute preferences into its optimization process. We further benchmarked KomoTrip against state-of-the-art TOPTW heuristics on the public Solomon dataset, where it demonstrated competitive profit values, particularly for a larger number of days (tours), and consistently achieved superior runtime performance.

Given a graph whose nodes and edges are associated with a profit, a visiting (or traversing) time and an admittance time window, the Mixed Team Orienteering Problem with Time Windows (MTOPTW) seeks for a specific number of walks spanning a subset of nodes and edges of the graph so as to maximize the overall collected profit. The visit of the included nodes and edges should take place within their respective time window and the overall duration of each walk should be below a certain threshold. In this paper we introduce the MTOPTW, which can be used for modeling a realistic variant of the Tourist Trip Design Problem where the objective is the derivation of near-optimal multiple-day itineraries for tourists visiting a destination which features several points of interest (POIs) and scenic routes. Since the MTOPTW is a NP-hard problem, we propose the first metaheuristic approaches to tackle it. The effectiveness of our algorithms is validated through a number of experiments on POI and scenic route sets compiled from the city of Athens (Greece).

Mobile tourist guides evolve towards automated personalized tour planning devices. The contribution of this article is to put forward a combined artificial intelligence and metaheuristic approach to solve tourist trip design problems (TTDP). The approach enables fast decision support for tourists on small footprint mobile devices. The orienteering problem, which originates in the operational research literature, is used as a starting point for modelling the TTDP. The problem involves a set of possible locations having a score and the objective is to maximize the total score of the visited locations, while keeping the total time (or distance) below the available time budget. The score of a location represents the interest of a tourist in that location. Scores are calculated using the vector space model, which is a well-known technique from the field of information retrieval. The TTDP is solved using a guided local search metaheuristic. In order to compare the performance of this approach with an algorithm that appeared in the literature, both are applied to a real data set from the city of Ghent. A collection of tourist points of interest with descriptions was indexed and subsequently queried with popular interests, which resulted in a test set of TTDPs. The approach presented in this article turns out to be faster and produces solutions of better quality.

The shift from mass tourism to more personalized travel denotes great importance in the construction of tourist itineraries. Given the negative impacts of transport and tourism on the environment, sustainability criteria play an important role. The Tourist Trip Design Problem is related to the design of itineraries for tourists. Planning is complex in tourist regions of developing countries where the information associated with tourist activities is difficult to access, vague, and incomplete. With this information, tourists must plan their trip, and the conditions and limitations they establish for it are flexible and imprecise. Fuzzy optimization can address problems with this type of information and constraints. Therefore, in this paper, an analysis of the tourism supply chain is carried out, taking as a case study the Department of Sucre on Colombia's Caribbean coast. A Multiconstraint Multimodal Team Orienteering Problem with Time Windows and fuzzy constraints is developed to model the tourism trip design problem that maximizes profit and minimizes CO2 emissions. The model is tested using datasets from the literature and the real world. The results demonstrate consistency with the fuzzy approach and generate a set of low-emission solutions.

This paper presents a new approach to a recommendation of learning activities adapted to the spatial and temporal context of each mobile learner. Indeed, the path traveled by the user during a field trip can be guided using the technique of passive collaborative filtering. This recommendation is based on the ACO (Ant Colony Optimization) algorithm, which represents a good model for swarm intelligence. For this reason, the structure of our mobile scenario is described as a graph where POIs (Point Of Interest) are represented by nodes and the arcs indicate the probability of moving between them. This recommendation system allows the orchestration of mobile learning according to the geographical location of learners and the historical of their activities. Our contribution is devised in three parts: (1) the creation of a mobile learning scenario based on POIs, (2) the adaptation of the ACO algorithm for the orchestration of paths taken by learners, and (3) the development of a recommender system that helps learners to better choose their paths during the field trip. Index Terms—Mobile learning, field trip scenario, POI, orchestration of mobile learning activities, recommendation system, passive collaborative filtering, ACO algorithm.

Recommender Systems (RS) have been applied in several scenarios due to their ability to satisfy the user's interest. Traditionally, they have been applied in scenarios such as entertainment and e-commerce, and nowadays, in Location Based Social Network (LBSN) to recommend points-of-interest (POIs). Despite the advances in traditional scenarios, there is an opportunity for improvements in POI domains. For this scenario, it is necessary to consider the geography influence of POIs. However, we observe the main proposals are not able to achieve satisfactory results. In this work, we open a new research perspective in POI Recommendation area, proposing a post-processing approach that can be used with any RS. Basically, we measure the activity level of users in different subareas of a city and use it to re-order the POIs retrieved by a RS. We evaluate our proposal considering six recommender systems and three datasets from Yelp achieving gains up to 15% of precision.

In this digital era, the recommendation system is an important tool in helping users to make decisions, including choosing tourist destinations. This study aims to implement the K-Nearest Neighbors (k-NN) algorithm in determining the recommendation pattern for tourist destinations. The k-NN algorithm was chosen because of its ability to classify based on the closest neighbor of a point, so that it can obtain recommendation patterns that are more personal and relevant to user preferences. This implementation method involves data collection, data preparation, model building, classification process, model evaluation, and finally implementation of the model in a recommendation system. The results of this study indicate that a recommendation system based on the k-NN algorithm is able to provide suggestions for tourist destinations that are more in line with user preferences, thus potentially increasing user satisfaction and making a positive contribution to the tourism industry. However, the effectiveness of this model is highly dependent on the quality and quantity of data used. Therefore, good data collection and preparation is very important in the implementation of this algorithm. This research opens opportunities for further research and development in the field of tourist destination recommendation systems using other algorithms or combining several algorithms to obtain more optimal results. The effectiveness of this model is highly dependent on the quality and quantity of data used. Therefore, good data collection and preparation is essential in the implementation of this algorithm. This research opens up opportunities for further research and development in the field of tourist destination recommendation systems by using other algorithms or combining several algorithms to get more optimal results. The effectiveness of this model is highly dependent on the quality and quantity of data used. Therefore, good data collection and preparation is very important in the implementation of this algorithm. This research opens up opportunities for further research and development in the field of tourist destination recommendation systems by using other algorithms or combining several algorithms to get more optimal results.

The regular language-constrained orienteering problem with time windows

We introduce a new variation of the Orienteering Problem (OP), the Minimum-Maximum Category Constraints Orienteering Problem with Time Windows. In the Orienteering Problem we seek to determine a path from node \( S \) to node \( T \) in a weighted graph where each node has a score. The total weight of the path must not exceed a predetermined budget and the goal is to maximize the total score. In this variation, each Activity is associated with a category and the final solution is required to contain at least a minimum and at most a maximum of specific categories. This variation better captures the problem of tourists visiting cities. For example, the tourists can decide to visit exactly one restaurant at a specific time window and at least one park. We present a Replace Local Search and an Iterated Local Search which utilizes Stochastic Gradient Descent to identify the tightness of the constraints. We perform exhaustive experimental evaluation of our results against state of the art implementations for the unconstrained problem and examine how it performs against increasingly more restricting settings.

Point-Of-Interest (POI) Recommender Systems (RSs) have huge importance in Location Based Social Networks (LBSNs) because of its effectiveness in assisting users to explore personalized locations. It also assists the LBSN providers to increase their revenue through scrutinized advertisements or services according to specific locations. For the effectiveness and accuracy of POI RSs, so many additional information such as Transition Contexts (e.g., geographical distance, time interval), Dynamic Contexts (e.g., time of the day, companion, season), and Static Contexts (e.g., POI type, features) have to be integrated along with the check-in data. The high impact of this additional information distinguishes POI recommendation approaches from other RS approaches. To address the challenges of the varying influences of the user’s current contexts and the transition contexts (arises from users past task to current task) on recommendation, a Gated Recurrent Unit (GRU) architecture is proposed. It is capable of handling the effect of each category of contexts separately. The main part of the proposed Context-Category Specific Sequence Aware POI RS (CCS-POI-RS) is a Multi-GRU (MGRU), which has two added gates for handling the influences of both dynamic contexts and transition contexts. Experiments on Gowalla and Foursquare check-in data set reveal the significance of MGRU architecture through the comparison with the other state of the art GRU architectures.

In the Team Orienteering Problem with Time Windows (TOPTW), a variant of the Vehicle Routing Problem with Profits, a set of locations is given, each associated with a profit, a visiting time and a time window. The aim is to maximize the overall profit collected by a number of routes, while the duration of each route must not exceed a given time budget. TOPTW is NP-hard and is typically used to model the Tourist Trip Design Problem. The latter deals with deriving near optimal multiple-day tours for tourists visiting a destination with several points of interest (POIs). The most efficient known heuristic approach to TOPTW which yields the best solution quality versus execution time, is based on Iterated Local Search (ILS). However, the ILS algorithm treats each node separately, hence it tends to overlook highly profitable areas of nodes situated far from the current solution, considering them too time-expensive to visit. We propose two cluster-based extensions to ILS addressing the aforementioned weakness by grouping nodes on separate clusters (based on geographical criteria), thereby making visits to such nodes more attractive. Our approaches improve ILS with respect to solutions quality and execution time as evidenced by experimental tests exercised on both existing and new TTDP-oriented benchmark instances.

With the massive growth of the internet, a new paradigm of recommender systems (RS's) is introduced in various real time applications. In the research for better RS's, especially in the travel domain, the evolution of location-based social networks have helped RS's to understand the changing interests of users. In this article, the authors present a new travel RS employed on the mobile device to generate personalized travel planning comprising of multiple Point of Interests (POIs). The recommended personalized list of travel locations will be predicted by generating a heat map of already visited POIs and the highly relevant POIs will be selected for recommendation as destinations. To enhance the recommendation quality, this article exploits the temporal features for increased user visits. A personalized travel plan is recommended to the user based on the user selected POIs and the proposed travel RS is experimentally evaluated with the real-time large-scale dataset. The obtained results of the developed RS are found to be proficient by means of improved diversity and accuracy of generated recommendations.

In this paper, an algorithm for making travel itinerary using traveling salesman problem (TSP) and k-means clustering technique is proposed. We employ the algorithm to develop a web based application that can help travelers to plan their travel itinerary. The developed application should be able to provide an optimal itinerary recommendation in terms of distance and travel time. We use initial assumption that the traveler has determined all the tourist destinations he/she wants to visit and also the number of days he/she will stay in the region. Our approach consists of two steps, macro grouping using k-means and micro tour arrangement using TSP. Yogyakarta city, one of the tourist city in Indonesia, is used as an example to illustrate how the proposed algorithm can help travelers make their itinerary. This approach works well in small to medium number points of interest. However, the application still need many improvements such as to make it run faster and to handle the additional constraints that exist when creating an itinerary.

Recommender systems can help web users find more relevant content, improve their online experience, and support them in the discovery of new Points-of-Interest (POI). Yet, challenges persist in dealing with the cold-start problem and in recommendation explainability. To address these, we have created ScrollyPOI, an interactive POI recommender system based on Data Humanism principles. Utilizing scrollytelling, we address the cold-start problem by engaging users in reflecting on previous positive experiences. Additionally, ScrollyPOI enhances explainability through input and output explanations. The system uses stacked bar charts and word clouds to explain how user preferences inform recommendations (input). Finally, ScrollyPOI employs a multi-layered approach to explain why specific POIs are recommended (output). We have evaluated ScrollyPOI’s interface and experience through a preliminary study, highlighting its potential for transparent explanations in the POI recommendation domain. Our findings underscore ScrollyPOI’s efficacy in collecting preferences and enhancing recommendation transparency, positioning it as a platform for studying explainability goals in the POI domain.

The point-of-interest (POI) recommender exploits check-in information from location-based social networks (LBSNs) to recommend POIs that match user preferences. The preferences of users vary with region/locality, consumption type, and co-consumers. For instance, the places preferred for friends may be different from the places preferred for family, and the places preferred in one locality may be different from the places in another locality. These dynamic preferences can be crucial for an efficient recommender system. A locality consists of different preference trends, such as “known for food and recreational sites.” In real-world, different sets of users might be attracted toward different preference trends, and some users' preferences might overlap across multiple preference sets. Hence, an efficient aggregation of locality preferences is essential for the recommender systems. Many existing studies simply group items by their category and use simple collaborative filtering (CF) for recommendations, however, such techniques cannot efficiently handle the aggregated locality preferences. We propose a hierarchical recommendation model termed Hierarchical Contextual Location Recommendation System (HiRecS) that formulates users' preferences as a hierarchical structure and models the locality trend using aggregated hierarchy. For a locality, the root of hierarchy contains preferred k items from a set of visitors, and the subsequent levels contain preference wise subsets of those items. We also present a hierarchy aggregation technique to aggregate the hierarchical preferences from a similar set of users. The aggregated hierarchy is then contextually exploited for POI sequence recommendation. The core contributions of this article are: 1) it formulates the locality trends as hierarchical structures, presents a hierarchy aggregation technique, and models the personalized POI preferences with the aggregated hierarchy; 2) it exploits the aggregated popular trends to generate contextual POI sequence recommendation; and 3) it extensively evaluates the proposed model with two real-world data sets and demonstrates a significant performance gain of 0.006-5.91 on diversity metrics, 0.0349-17.51 on displacement metrics, and 0.114-0.289 on normalized discounted cumulative gain (NDCG) metrics, when compared to several baselines and relevant studies.