ABSTRACT This paper presents SmartTur+ECO, a system for tourist attractions that integrates conversational recommendation systems (CRS), virtual reality (VR) and gamification. It efficiently gathers trip data and elicits user preferences through interactions with a chatbot, addressing the user‐cold start problem common in tourism. Using an exploit‐explore approach with sentence embeddings from POI descriptions, SmartTur+ECO provides precise and personalized recommendations organized into multi‐day plans, considering opening hours and locations. VR integration offers immersive pre‐trip experiences, helping users make informed decisions. Gamification elements like mini‐games and rewards are designed to foster user engagement and motivation. SmartTur+ECO's modular and scalable architecture ensures compatibility with various devices and allows for the easy integration of new algorithms and data sources, with the aim of supporting recommendation processes. This innovative combination sets new standards in tourism by making travel planning more efficient, entertaining and meaningful.

Abstract E-learning has transformed the educational landscape, particularly in higher education, by offering flexible, scalable, and often more accessible learning environments. Lecture recordings, in particular, have become a widely used resource, offering students the ability to revisit class content at their own pace. However, the sheer volume and length of these recordings can make it difficult for learners to locate specific types of instructional content efficiently. This paper presents a hierarchical multimodal approach to segment and classify lecture recordings based on the nature of the teaching activity taking place. The proposed method integrates audio processing techniques and natural language understanding models to distinguish between various communicative functions, such as content delivery, task explanation, or organizational announcements. By leveraging both acoustic and textual cues, the system enables more effective navigation through educational videos, facilitating targeted access to relevant material. Experimental results demonstrate high overall accuracy and notable improvements over existing approaches, especially in identifying structured instructional discourse. Nonetheless, challenges persist in detecting informal or less clearly defined interactions. This work aims to enhance the usability of recorded lectures and support more personalized and efficient learning experiences.

Research in planning has sought to broaden the scope of planning problems by incorporating numeric parameters into action descriptions to condition both continuous and discrete change. Focusing on the latter, this work studies the problem of numeric planning with control variables, a reformulation of actions with infinite domain parameters. To tackle the challenge of handling an infinite decision space driven by control variables, we incorporate sampling into a forward state-space search. The resulting search framework (1) partially expands nodes by sampling their successors and (2) implements a re-expansion strategy to sample additional successors if a node shows promise in future evaluations. We perform a deep probe into this concept that materializes into a new algorithm called Sampling Greedy Best-First Search (S-GBFS). Our empirical evaluation of S-GBFS across various domains shows significant improvements over existing planning approaches.

In automated planning, control parameters extend standard action representations through the introduction of continuous numeric decision variables. Existing state-of-the-art approaches have primarily handled control parameters as embedded constraints alongside other temporal and numeric restrictions, and thus have implicitly treated them as additional constraints rather than as decision points in the search space. In this paper, we propose an efficient alternative that explicitly handles control parameters as true decision points within a systematic search scheme. We develop a best-first, heuristic search algorithm that operates over infinite decision spaces defined by control parameters and prove a notion of completeness in the limit under certain conditions. Our algorithm leverages the concept of delayed partial expansion, where a state is not fully expanded but instead incrementally expands a subset of its successors. Our results demonstrate that this novel search algorithm is a competitive alternative to existing approaches for solving planning problems involving control parameters.

In this article, we propose a centralized Multi-Agent Learning framework for learning a policy that models the simultaneous behavior of multiple agents that need to coordinate to solve a certain task. Centralized approaches often suffer from the explosion of an action space that is defined by all possible combinations of individual actions, known as joint actions. Our approach addresses the coordination problem via a sequential abstraction, which overcomes the scalability problems typical to centralized methods. It introduces a meta-agent, called supervisor, which abstracts joint actions as sequential assignments of actions to each agent. This sequential abstraction not only simplifies the centralized joint action space but also enhances the framework's scalability and efficiency. Our experimental results demonstrate that the proposed approach successfully coordinates agents across a variety of Multi-Agent Learning environments of diverse sizes.

The recognition of activities performed by humans is crucial in human-robot interaction. However, relying solely on the assumption of rational human behaviour may be flawed given the large influence of individual preferences on decision-making. This paper proposes a novel approach to understanding human behaviour by examining how individuals select actions to tackle various tasks. Central to this approach is the concept of a Relational Decision Tree, which encapsulates human behaviour patterns. Then, when a new planning problem is posed, this relational decision tree is used to generate a plan that follows the learned human behaviour. This methodology enables proactive anticipation of individual needs and tailored responses. Validation across five diverse domains and several behaviours demonstrates the efficacy of this method.

The recognition of activities performed by humans is crucial in human-robot interaction. However, assuming humans always follow rational behaviour in executing activities may not be accurate since individual preferences influence their decision-making. This paper proposes a method for learning human behaviour that involves capturing how humans select actions to solve problems. This behaviour is represented by a Relational Decision Tree. We define two sets of features that can be automatically extracted from the planning domain. A behaviour library is created and used to identify the behaviour followed by a person when executing a plan in a new situation. This approach allows to anticipate the person’s needs and act accordingly. The method was tested in three different domains, showing its validity.

The aim of this work is to explain the observed behaviour of a hybrid system (HS). The explanation problem is cast as finding a trajectory of the HS that matches some observations. By using the formalism of hybrid automata (HA), we characterize the explanations as the language of a network of HA that comprises one automaton for the HS and another one for the observations, thus restricting the behaviour of the HS exclusively to trajectories that explain the observations. We observe that this problem corresponds to a reachability problem in model-checking, but that state-of-the-art model checkers struggle to find concrete trajectories. To overcome this issue we provide a formal mapping from HA to PDDL+ and show how to use an off-the-shelf automated planner. An experimental analysis over domains with piece-wise constant, linear and nonlinear dynamics reveals that the proposed PDDL+ approach is much more efficient than solving directly the explanation problem with model-checking solvers.

There are many models that attempt to predict vehicular speed in urban and interurban roads, the noise pollution caused by traffic in cities, or even the traffic flow based on historical data from cameras or from people's mobile phones. Such information can be useful for administration authorities, and for researchers attempting to improve the living conditions of citizens. In this context, the aim of the present study is to design a model capable of predicting the traffic flow in the city of Valencia, Spain, based on data collected by electromagnetic loops distributed throughout the city. With a good traffic prediction, it will be possible to foresee possible traffic jams, and also to trigger countermeasures to mitigate them. Therefore, two models based on two recurrent neural networks of Long Short-Term Memory (LSTM) type have been designed to predict the traffic flow in the different streets of Valencia at the different hours of the day. We also study the influence of the specific characteristics used on the accuracy of the model. The results of our experiments show that, despite the high heterogeneity in terms of per-street traffic behaviour, it is possible to reach useful prediction models with low errors.

A declarative action model is a compact representation of the state transitions of dynamic systems that generalizes over world objects. The specification of declarative action models is often a complex hand-crafted task. In this paper we formulate declarative action models via state constraints, and present the learning of such models as a combinatorial search. The comprehensive framework presented here allows us to connect the learning of declarative action models to well-known problem solving tasks. In addition, our framework allows us to characterize the existing work in the literature according to four dimensions: (1) the target action models, in terms of the state transitions they define; (2) the available learning examples; (3) the functions used to guide the learning process, and to evaluate the quality of the learned action models; (4) the learning algorithm. Last, the paper lists relevant successful applications of the learning of declarative actions models and discusses some open challenges with the aim of encouraging future research work.

This paper introduces the Temporal Inference Problem (TIP), a general formulation for a family of inference problems that reason about the past, present or future state of some observed agent. A TIP builds on the models of an actor and of an observer. Observations of the actor are gathered at arbitrary times and a TIP encodes hypothesis on unobserved segments of the actor's trajectory. Regarding the last observation as the present time, a TIP enables to hypothesize about the past trajectory, future trajectory or current state of the actor. We use LTL as a language for expressing hypotheses and reduce a TIP to a planning problem which is solved with an off-the-shelf classical planner. The output of the TIP is the most likely hypothesis, the minimal cost trajectory under the assumption that the actor is rational. Our proposal is evaluated on a wide range of TIP instances defined over different planning domains.

Research on AI is more and more focusing towards explainable technology that accounts for the outcomes of programs and products. One important aspect in this direction is the ability to recognize the behaviour patterns of our application in order to make sensible and informed decisions. In this work, we aim to uncover the behaviour of a planning agent by means of Behaviour Trees (BTs), a flexible and controllable mathematical model of plan execution used to describe flows between tasks in a modular fashion. We analyze the behaviour of a planning agent when solving a simple logistics problem by comparing the agent’s plan with the plans resulting from various BTs, each representing a different behaviour. We propose different distance metrics as a similarity measurement between plans and we evaluate their accuracy at identifying similar behaviours.

Exploring the mechanism about users' emotion dynamics towards social events and further predicting their future emotions have attracted great attention to the researchers. Despite the concreteness of the online expressions in written form, it remains unpredictable which kinds of emotions will be expressed in individual messages of Twitter users influenced by his/her friends. To investigate this, we perform an investigation on observing emotions unfolding in a consecutive sequence of tweets for a particular user based on his/her past history. In this paper, we propose an Emotion-based User Sequential Influence Model (E-USIM) on given a set of tweets related with some events (identified by the usage of a hashtag), determines how those sentiments will be distributed on behalf of a person within a conversation. We then apply the developed model to predict users' future emotions by combing of personal and interpersonal influence.

Recently, social media has been considered the fastest medium for information broadcasting and sharing. Considering the wide range of applications such as viral marketing, political campaigns, social advertisement, and so on, influencing characteristics of users or tweets have attracted several researchers. It is observed from various studies that influential messages or users create a high impact on a social ecosystem. In this study, we assume that public opinion on a social issue on Twitter carries a certain degree of emotion, and there is an emotion flow underneath the Twitter network. In this article, we investigate social dynamics of emotion present in users’ opinions and attempt to understand (i) changing characteristics of users’ emotions toward a social issue over time, (ii) influence of public emotions on individuals’ emotions, (iii) cause of changing opinion by social factors, and so on. We study users’ emotion dynamics over a collection of 17.65M tweets with 69.36K users and observe 63% of the users are likely to change their emotional state against the topic into their subsequent tweets. Tweets were coming from the member community shows higher influencing capability than the other community sources. It is also observed that retweets influence users more than hashtags, mentions, and replies.

Exploring the mechanism about users' emotion dynamics towards social events and further predicting their future emotions have attracted great attention to the researchers. One of the unexplored components of human communication found online in written form is an emotional expression. However, despite the concreteness of the online expressions in written form, it remains unpredictable which kinds of emotions will be expressed in individual messages of Twitter users. To investigate this, we perform an investigation on observing emotions unfolding in a consecutive sequence of tweets for a particular user based on his/her past history. In this paper, we propose a method on given a set of tweets related with some events (identified by the usage of a hashtag), determines how those sentiments will be distributed on behalf of a person within a conversation. We present the Hidden Markov Model (HMM) to understand the nature of emotion dynamics in Twitter messages.

This paper describes e-Tourism2.0, a web-based recommendation and planning system for tourism activities that takes into account the preferences that define the travel style of the user. e-Tourism2.0 features a recommender system with access to various web services in order to obtain updated information about locations, monuments, opening hours, or transportation modes. The planning system of e-Tourism2.0 models the taste and travel style preferences of the user and creates a planning problem which is later solved by a planner, returning a personalized plan (agenda) for the tourist. e-Tourism2.0 contributes with a special module that calculates the recommendable duration of a visit for a user and the modeling of preferences into a planning problem.

When two or more self-interested agents put their plans to execution in the same environment, conflicts may arise as a consequence, for instance, of a common utilization of resources. In this case, an agent can postpone the execution of a particular action, if this punctually solves the conflict, or it can resort to execute a different plan if the agent's payoff significantly diminishes due to the action deferral. In this paper, we present a game-theoretic approach to non-cooperative planning that helps predict before execution what plan schedules agents will adopt so that the set of strategies of all agents constitute a Nash equilibrium. We perform some experiments and discuss the solutions obtained with our game-theoretical approach, analyzing how the conflicts between the plans determine the strategic behavior of the agents.

In this paper we present a temporal planning approach for handling problems with deadlines. The model relies on the extraction of temporal landmarks from the problem and the construction of a landmarks graph as a skeleton of the solution plan. Partial plans in the search tree that are not compliant with the information comprised in this graph are pruned. We introduce a novel search scheme that builds a landmarks graph in each tree node and which notably improves the rate of detection of unsolvable problems.

This paper presents a novel multi-agent reactive execution model that keeps track of the execution of an agent to recover from incoming failures. It is a domain-independent execution model, which can be exploited in any planning control application, embedded into a more general multi-agent planning framework. The multi-agent reactive execution model provides a mechanism allowing an agent to respond to failures that prevent completion of a task when another agent is not able to repair the failure by itself. The model exploits the reactive planning capabilities of agents to come up with a solution at runtime, thus preventing agents from having to resort to replanning. We show the application of the proposed model for the control of multiple autonomous space vehicles.

Authoring tools for building intelligent educational systems must provide support to ensure flexibility, adaptability of content to the user profile, reusability and sharing of learning objects. These facilities are essential to develop automated decision processes for providing course compositions tailored to the specific characteristics of each individual learner. We present a LOM-compliant learning approach that uses an automated planning process to create personalised learning courses while giving special attention to the development of reusable learning objects.

Deadline constraints have been recently introduced in PDDL3.0. The results obtained in the constraints domains in the last Planning Competition show that planners are not yet fully competitive. When dealing with deadline constraints the number of feasible solutions for a problem is reduced and thus it is specially relevant the ability to detect unsolvability. In this paper we present a new approach, based on the use of temporal landmarks, for the detection of unsolvable temporal planning problems.

In real world we have to deal with changing situations which may partially or entirely invalidate an executable plan. Current strategies for plan repair are basically aimed at solving problems where regular minor modifications in the initial or goal state occur in the plan. In this paper, we propose a new repair technique that identifies where the problem is located, which part of the plan needs to be repaired and it fixes the affected part of the plan. Our technique is capable to tackle any type of failure or modification in the plan.

The aim of educational systems is to design a sequence of learning objects on a set of topics tailored to the learner's goals and individual properties. However, some of the main difficulties actual educational systems have to face is the generation of learning routes for multiple learners, the lack of an explicit management of time and resources or the synchronization of group activities. We claim that AI planning provides the necessary technology to address all these missing issues in actual e-learning environments, and this paper elaborates in this direction.

e-Tourism is a tourist recommendation and planning application to assist users on the organization of a leisure and tourist agenda. First, a recommender system offers the user a list of the city places that are likely of interest to the user. This list takes into account the user demographic classification, the user likes in former trips and the preferences for the current visit. Second, a planning module schedules the list of recommended places according to their temporal characteristics as well as the user restrictions; that is the planning system determines how and when to perform the recommended activities. This is a very relevant feature that most recommender systems lack as it allows the user to have the list of recommended activities organized as an agenda, i.e. to have a totally executable plan.

The issue of temporal reasoning in an online real-time expert system in the field of process control is addressed. An alarm filtering problem is described, and the integration of a point-based time map manager (TMM) with an inference engine, to manage the temporal constraints, is proposed. The TMM manages only imprecise future time points and must comply with the constant modification of the current time and thus with the removal of time points that are moved to the past. Temporal restrictions in the left-hand sides of rules are also described, together with the need for management of pending queries that may receive an answer when the uncertainty pervading the future is reduced.>