The contribution of ANTIMOINE projects concernes «Digital Humanities« because it focuses on the study of innovative tools for participatory construction and dissimination of heritage ; the design of these tools is based on three areas : linguistics, data science and virtual reality.
ANTIMOINE project puts forward a new vision of tools essential for the knowledge building related to situated human activities (territorial anthropology), knowledge built up out of heritage objects and their interpretation (reading). People make society to live. They make history. Heritage is the trace of their production over time. The challenge of the future is to read this heritage to identify what sets them apart and what brings them together. <br /> <br />Heritage constitutes a complex system in the sense that understanding an element requires understanding many other elements. Reading heritage and responding to questions of a user cannot be done unless building an interpretation context consisting of interlinked objects related to these questions. This construction results in the process of a heritage database (in which the questioning, often vague, gradually becomes clear). <br /> <br />At present, the heritage information systems work mainly on the basis of keywords with very limited consideration of a pre-defined semantics. Moreover, their databases are enriched neither automatically nor semi-automatically. <br /> <br />The break proposed by the ANTIMOINE project with regards to the state of the art is twofold: on the one hand, it offers to get linguistics, data science and virtual reality cooperate to structure the heritage database and create new relationships between the data. On the other hand, it intends to participate in the creation of new formal models of heritage to effectively control the structuring and creation. The latter ambition is based in particular on the possibility for users to interact with these data and models. <br /> <br />The argument of the ANTIMOINE project is that this break can only be done by introducing meaning within the heritage information systems, making it easier to read heritage. The originality of ANTIMOINE is to seek to make cooperate three different semantic levels: <br /> <br />- meaning through text analysis <br />- meaning through data analysis <br />- meaning through action analysis
ANTIMOINE tackles three academic barriers and a technological limitation:
- Linguistic barreer
- Data mining barreer
- Virtual reality barreer
- Technological barreer
To overcome these obstacles we propose to adopt an interdisciplinary approach based on the following solutions (theories and methods):
- argumentative probabilities semantics (linguistics) for the introduction of meaning by analysing texts
- frequent patterns analysis (data mining) for the introduction of meaning by analysing data
- enaction (virtual reality) for the introduction of meaning by analysing users’ actions based on sensory-motor loops users / information system
The whole system is based on four models : .
- the ontological model (concept of heritage, association between them)
- the data model (the annoted heritage database)
- the data mining mode (a dual representation based on frequent patterns)l
- the presentation model (a 3d animated representation of the heritage database)
One of the central points of the project will involve the study of control loops (illustrated in the figure above) essential to control the production of data and their presentation:
- Control loop for data production
- Control loop for human-system interface
- Control loop human- meta (where the traces of the user's activities influence the extraction of frequent pattern)
The argument of the ANTIMOINE project (tripartite introduction of meaning in heritage databases) will be evaluated through experiments conducted using a prototype built incrementally from software modules developed by the different partners and integrated into the information system of the partner company Topic-Topos.
Three evaluation scenarios will be considered. They use the same database, but correspond to different uses (museum, education and tourism).
The project ANTIMOINE began in March 2014.
Current studies are mainly a presentation of each partner bases:
- a presentation of the structure of the database as it exists. This presentation helped to highlight the different information that characterize the heritage and the relationships that bind these elements. This information is important because they determine not only the structure of linguistic concepts produie but also the extraction of frequent patterns algorithms.
- an overview of the concepts used in the linguistic framework (semantic of possible argumentative), the methodology used to get from texts concepts and associations between these concepts. Work has been done to guide the SPA to produce data for the project.
Several perspectives are possible (they will be re-evaluated as the project progresses):
From a scientific point of view:
- Consider new modes of interaction
- Study the interaction between data mining and human-machine interface
From a technical point of view:
- Integrate technical bricks in the operating income of the business partner Topic-Topos
From an economic point of view:
- Allow the Topic-Topos company to offer, from the new technical building blocks of new services based on research information facilitated by the use of these new interfaces.
- Consider a spin-off Topic-creation Topos specializes in creating content for the heritage information systems developed from the acquired ANTIMONY project.
- Provide tools for editing-author of the ontology based on the results of data mining.
actually nothing (due to the launch date of the project)
ANTIMOINE is a project focused on the tools needed to conduct an anthropological analysis of territories from cultural heritage data.
On the application point of view areas are education, tourism, land, territories project management ...
From a scientific point of view the aim of the project is the introduction of meaning in legacy information systems so as to provide, in response to a user inquiry, a favorable environment cultural heritage interpretation. This context consists of a set of cultural heritage objects linked by associations with a semantic feature heritage. This context, not defined a priori is obtained in the course of a database heritage.
To achieve this goal ANTIMOINE adopts an interdisciplinary approach involving linguistics, data mining and virtual reality.
If each partner brings its own lock, fifth lock, transverse emerges: the cooperation between processes specific to each discipline. Cooperation allowing on one hand the introduction of the second direction to reduce the complexity.
To solve these locks we rely on three theories: Possible Argmentatifs Semantics (linguistics), the analysis of frequent patterns (data mining) and enaction (virtual reality).
This work will be illustrated and evaluated through a prototype operating a real database. This prototype will be developed incrementally and will allow the implementation of three scenarios: education, tourism and museum .. With each scenario a different hardware platform: desktop, digital tablet, immersive device.
To get to overcome these obstacles ANTIMOINE relies on a consortium of four partners: the laboratory Lab-STICC - ENIB (Virtual Reality), LINA-COD (data mining) and CoDIRe (linguistic) and society Topic-Topos (database administration and software integration).
The work to be performed is divided into five tasks:
Task 1 : coordination
Task 2 : semantic analysis of cultural heritage . analysis of the semantics of inheritance. It is to propose and use the tools from texts and other data used to create semantic models of heritage, ie The semantic features (concepts) and associations between these rules.
Task 3 : data mining . It is to study and develop tools to discover groups of cultural heritage objects and associations between these groups not previously explained and exploited to amend the semantic models.
Task 4 : enactive interface .It is to propose a method of natural and immersive interaction can self-organize in order to provide a context for cultural heritage interpretation in line with the questioning of the user.
Task 5 : integration . This task has several objectives: (1) ensure the exchange of data between different modules and their synchronization, (2) ensure the integration of software modules developed by the partners and (3) ensure the system's portability between different platforms material forms
Monsieur Eric MAISEL (Laboratoire des Sciences et Technologies de l'Information de la Communication et de la Connaissance)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LINA Laboratoire d'Informatique de Nantes Atlantique
CoDiRe Construction Discursive des Représentations linguistiques et culturelles
Lab-STICC-ENIB Laboratoire des Sciences et Technologies de l'Information de la Communication et de la Connaissance
Help of the ANR 459,342 euros
Beginning and duration of the scientific project: February 2014 - 42 Months