Personalized and context-aware experiences
| We present semantics-based mechanisms that aim to promote reflection on cultural heritage by means of dates (historical events or annual commemorations), owing to their connections to a collection of items and to the visitors’ interests. We argue that links to specific dates can trigger curiosity, increase retention and guide visitors around the venue following new appealing narratives in subsequent visits. The proposal has been evaluated in a pilot study on the collection of the Archaeological Museum of Tripoli (Greece)... | |
| This paper is a first step towards identifying the links between the characteristics of gaze behaviour and visitor preferences in a museum. In the long term, the real-time analysis of visitors' gaze should allow a fine estimation of their interest for the different artworks exhibited and should replace the fastidious and time-consuming elicitation of preferences commonly used in traditional recommender systems. To study these links, we carried out a user study at the Nancy Museum of Fine Arts in the North-East of France... | |
| Recommender systems are a flourishing domain in computer science for almost 30 years now. This rising popularity follows closely the number of data collected all around the world. Each and every internet user produces a huge amount of content during his lifetime. Recommender systems proactively help users to navigate these pieces of information by gathering, and selecting the items to users' needs. In this paper, we discuss the possibility and interest of applying our Multi-Objective Ant Colony System called AntRS to recommend items in different application domains... | |
| Biclustering plays a crucial role in many real world applications. Related to clustering, which groups similar rows in a matrix (data table), biclustering aims at simultaneously grouping similar rows and columns, i.e. to find submatrices which exhibit a correlation among their respective cells. There are many types of biclustering based on a similarity criterion. In this paper we are interested in constant-column (CC) biclustering... | |
| Presentation. | |
| This paper focuses on item recommendation for visitors in a museum within the framework of European Project CrossCult about cultural heritage. We present a theoretical research work about recommendation using biclustering. Our approach is based on biclustering using FCA and partition pattern structures. First, we recall a previous method of recommendation based on constant-column biclusters... | |
| This paper presents our work about mining visitor trajectories in Hecht Museum (Haifa, Israel), within the framework of CrossCult European Project about cultural heritage. We present a theoretical and practical research work about the characterization of visitor trajectories and the mining of these trajectories as sequences. The mining process is based on two approaches in the framework of FCA, namely the mining of rare general subsequences and the mining of frequent contiguous subsequences... | |
| This paper presents our work about mining visitor trajectories, within the framework of CrossCult European Project about cultural heritage. We present a theoretical and practical research work about the characterization of visitor trajectories and the mining of these trajectories as sequences. The mining process is based on two approaches, namely the mining of subsequences without any constraint and the mining of... | |
| This paper describes how a surrogate model of the interrelations between different types of content in the same game can be used for level generation. Specifically, the model associates level structure and game rules with gameplay outcomes in a shooter game... | |
This paper introduces a surrogate model of gameplay that learns the mapping between different game facets, and applies it to a generative system which designs ew content in one of these facets. Focusing on the shooter game... | ||
| The question of representing emotion computationally remains largely unanswered: popular approaches require annotators to assign a magnitude (or a class) of some emotional dimension, while an alternative is to focus on the relationship between two or more options... | |
| his paper describes an iterative process for generating game levels by means of interlocking rooms evolved individually. The process is highly controllable by a human designer who can specify the entrances to this room as well as its size, its distribution of game objects and its architectural patterns. The small size of each room allows for computationally fast evaluations of several level qualities, but these rooms can be combined into a much larger game level. While the focus of this paper was on shooter game levels with two floors, the approach could be expanded to any generative process which is led by a designer. More ambitiously, the work could be led by a curator who can specify the details of a room and can then allow the generator to populate the digital version of this room with content particular to the intended play experience... | |
| This paper combines two popular divergent search approaches in evolutionary computation (novelty search and surprise search) to better explore the space of possible solutions. The hypothesis is that an evolutionary process that rewards both novel and surprising solutions will be able to handle deception in a better fashion and lead to more successful solutions faster. The results in maze navigation show that the combined novelty-surprise search is superior to individual algorithms, allowing the controller to better exploit its surroundings and be better aware of the context of its movement... | |
| This paper presents a two-step generative approach for game levels, providing with either a human or a computational designer a way to define a high-level overview of a map. Then the computational designer can populate the details of each cell in the high-level overview with game content. While the focus of this paper is on dungeons, which are popular for a number of game genres, the work could be expanded for generation of maps intended for urban gameplay. The high-level overview of the map, for example, could be made available for curators to allow them to define the experience that they would wish the generator to offer within their intended play locale... | |
| This poster presents how the locomotion of soft robots, which consist of different materials, is affected by different divergent search approaches. Findings indicate that novelty search and surprise search favor different robot morphologies, which in turn lead the robots to move and explore their surroundings in different ways... PDF link | |
| This paper introduces a computational model which can classify the gameplay outcomes of a game using only information on the initial state of the game rules and game level. The computational model uses convolutional neural networks to learn how game balance is affected by the level, represented as an image, and each team's weapon parameters. The deep learning approach shows the possibilities of a computational designer to be aware of the spatial context (i.e. game level) and parametric context (i.e. rules) when making a creative decision... | |
| There is a wide range of metadata standards for the documentation of museum related information, such as CIDOC-CRM; these standards focus on the description of distinct exhibits. In contract, there is a lack of standards for the digitization and documentation of the routes followed and information provided by museum guides. In this work we propose the notion of "narrative'', which can be used to model a guided museum visit... | |
| Offering personalised recommendations to visitors of a museum is a complex problem inherent to physical spaces. When at the same time specific applicative or museum objectives have to be taken into account, this becomes even more complicated. We introduce here a graph-based semantic recommender approach relying on ontological formalisation of knowledge about manipulated entities to solve the multi-dimensional recommendation problem encountered in museums... Available at | |
| The goal of this paper is to motivate the development and sketch an early design of the "Relevant dates" component of the CrossCult platform, which is intended to discover and recommend temporal-spatial relationships linking items or collections of cultural heritage to the current date. The selection of certain semantic resources is explained, and some usage examples in the contexts of pilot 2 and 3 are given... Available at | |
| This paper presents the webIcononscope tool for creative drawing, which allows users to draw simple icons composed of basic shapes and colors in order to represent abstract semantic concept. To complement the creativity of the human users attempting to create novel icons, several computational assistants provide suggestions... | |
| Wishing to connect cultural heritage, games and social networks, this work describes mini games to be used in CrossCult. For the purposes of supporting the museum visit, before, during and after, 5 games were designed for social networks to accomplish user profiling, to promote the museum and the application through social network dissemination, to introduce museum items and themes and to also function as visit souvenirs... | |
| Contrary to many application domains, recommending items within a museum is not only a question of preferences. Of course, the visitors expect suggestions that are likely to interest or please them. However, additional factors should be taken into account. Recent works use the visiting styles or the shortest distance between items to adapt the list of recommendations. But, as far as we know, no model of the literature aims at inferring in real time an holistic user model which includes variables such as the crowd tolerance, the distance tolerance, the expected user control, the fatigue, the congestion points... | |
| This paper applies a divergent evolutionary search method based on surprise to the constrained problem of generating balanced and efficient sets of weapons for the Unreal Tournament III shooter game. The proposed constrained surprise search algorithm ensures that pairs of weapons are sufficiently balanced and effective while also rewarding unexpected uses of these weapons during game simulations with artificial agents... | |
| This paper describes a search-based generative method which creates game levels by evolving the intended player experience rather than their spatial layout. The proposed approach evolves graphs where nodes representing player actions are linked to form one or more ways in which a mission can be completed ... | |
| This paper explores how an autonomous computational designer can create frames of tension which guide the procedural creation of levels and their soundscapes in a digital horror game. Using narrative concepts, the autonomous designer can describe an intended experience that the automated level generator must adhere to... | |
| his paper presents a playable demonstration of the Sonancia system, a multi-faceted content generator for 3D horror games, with the capability of generating levels and their corresponding soundscapes... | |
An EU project
This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement no 693150