J. Walsh and S. Erle. The epistomat: Generating consensus abour rdf ontologies and rules. Technical report, O’Reilly & Associates, Inc., London, UK, 2003.
————————————–

An ontology specifies the way that people interface with information. In computer science, an ontology is the attempt to formulate an exhaustive and rigorous conceptual schema within a given domain, a typically hierarchical data structure containing all the relevant entities and their relationships and rules (theorems, regulations) within that domain. (wikipedia).

The core idea of this paper is to use a predefined ontology defined in OWL as bootstrap for a world model to use in an interface. Then the paper focus on the idea that the connections and word of the ontology have to pass through a social process which will rate them and will give consent to them. They extract this principle from a game called Nomic. In this game players start with a predefined number of rules which progressively change during the game. The way new rules are applied is through a voting process.

In the author’s view the same system may be used to evolute the ontology used during the bootstrap of the interface. In my vision the action-series of the user may be the intentional subscription to the vote. This process is then very close to a social meaning process. In addition while the group give meaning to the ontology, also the peers of the group are learning how to assign these meaning though a democratic process. To highlight the informal learning process which takes place when different worldviews are negotiated by the consensus process.

Garage Cinema is doing research about Mobile Media Metadata. From their Internet site:

Our research is about making video a data type that humans and computers can create, access, process, reuse, and share according to descriptions of its semantic content and the principles of its syntactic construction. Our research aims to make this process as effortless as possible. Prof. Davis discussed the outlines of this agenda in an invited article for the 50th Anniversary Issue of the Communications of the ACM (Davis 1997).

[more…]

M. Davis, N. Good, and R. Sarvas. From context to content: Leveraging context from mobile media metadata. In SIGMOBILE, editor, MobiSys2004 proceedings, Boston, MA, USA, June 6–9 2004.
————————————-

This paper describes the Mobile Media Metadata System, which creates automaticaly a description of the content of pictures taken with SmartPhones from the context in which the picture was taken. Two features of the system are of particular interest: a) the annotation of the images is collaborative; b) the contextual metadata is built from the social interaction the user has with his/her peers.

In this context one of the most interesting aspect of this study is that a mean for an image can be taken from the way a picture is shared and reused in the group. For instance it is possible to infer the social context if the peers happens o be in the same cell at the same time and use this information in combination with others to know if the one of the peer is likely to be in the picture.

Again, is possible to infer spatial context from the cell id and time stamp of a picture. For a pedestrian, for example, is likely that two pictures represents the same area if these are taken with a couple of minutes of difference to one another.

Unfortunately the article does not go in details of the implemented algorithm. So it is not possible to know which strategies were used and how much to rely on those. Certainly they did not implement and machine learning algorithm.

Interesting link to the work of Toyama from which they took the idea of bootstraping contextual information from the header of the JPEG image.

Great brainstorm session today at CRAFT. We played attaching messages to a map of Lausanne. This can be an initial step in building an ontology of spatialised communication. Some ideas are here reported:

1. Maybe is possible to concatenate the messages so that the receiver can retrieve the follower once taken the one before. This can be achieved giving the user the possibility to set the spatial range at which the message can be retrieved.

2. A spatial wiki approach may be taken. We can give control to the participants to comment, rate and override messages.

3. We can use vector landmarks that shows the intentional direction of the emitter.

4. We can envisage an approaching alert to be used in a crowd scenario.

5. Gaming application.

From the evaluation of this activity I realised that it is difficult to relate the message with the position unless you can see the historical trace of the emitter. We have to consider that retrieving the message is asynchronous so by the time you retrieve people can have moved. From Nico’s descritpion:

…we tried to sketch out possible uses: recommending system (good/bad), help, threaded discussion, wiki-like (to reach a consensus), reminder system, spam/virus/breakdown, I’m late, anchors (like rave party system), synchronous/asychronous use, query (anybody wants to paly tennis here?), crowd scenario…

Other interesting comments about a possible criteria to categories the messages depending on the retrieving time (limited / unlimited) versus the receiver (one-many / range limited / everybody).

I presented the ideas I had to Pierre. The comments raised during the conversation concentrated on the definition of an initial ontology. He said that the kind of reasoning I am expecting is too advanced for machine learning. The kind of A.I. I am considering in the project is too far into neuro–network connections.

Something to do includes:
1. We have to know an initial ontology to start defining our algorithm;
2. We need to imagine what it could be;
3. What are the data we can take automatically from the system that mean something for the ontology?

WordNet® is an online lexical reference system whose design is inspired by current psycholinguistic theories of human lexical memory. English nouns, verbs, adjectives and adverbs are organized into synonym sets, each representing one underlying lexical concept. Different relations link the synonym sets.

Prof. Kuhn is a professor at the Institute for Geoinformatics at the University of Münster; and a lecturer (Dozent) at the Department of Geoinformation at the Technical University Vienna.

From his site:
The fundamental question underlying my research is how the meaning of geographic information can be mapped from one context to another. Typical contexts are human activities, professional disciplines, or natural languages. Contexts are ingrained in human minds, data models, user interfaces, business models, laws and regulations.

W. Kuhn. Why information science needs cognitive semantics – and what it has to offer in return, March 2003.
——————————

The author is well grounded on the semantics of geographic information. The notes describe the author’s point of view on cognitive semantics. His first approach is to give a background review of theoretical approach to the matter. Then he specifies that he likes the “action” approach, following which actions are seen as first–class entities. Pragmatics describes the rules for such actions. Semantics can be seen as a conventialised pragmatics. Syntax adds markers to help disambiguate expressions where context has already done so.

The author points to the concept of semantic interoperability to define a more sensible agreements across interfaces. This goes far beyond the simple syntactic agreement proposed by XML and the semantic web in general.

The author reports six tenets of Cognitive Semantics as defined by Peter G{“a}rdenfors:

1. Meaning is conceptualization in a cognitive system
2. Conceptual structures are embodied
3. Semantic elements are constructed from geometrical or topological structures
4. Cognitive models are primarily image-schematic. Image-schemas are transformed by metaphoric and metonymic operations.
5. Semantics is primary to syntax and partly determines it
6. Concepts show prototype effects

Regarding point 2., it is my idea that the ontology construction is actually embodied following the person interaction into the system.

Some core ideas are then listed by the author. Conceptualizations is defined as the cognitive activity of constituting our apprehension of the world. It is a dynamic concept. They are constructed. They can be constructed in multiple ways.

Features Bundles combine behavior with a list of attributes and the blending between them.

Image schemas are schematic and imagistic concepts which are abstracted from pre-cenceptual bodily experience, function as constituents of more complex notions, and provide the structure projected metaphorically to more abstract domains. For us to comprehend and reason about there must be pattern and order to our actions, perceptions, and conceptions. Image schemas can be subject to transformation: mappings from some aspects o schemas to others. maybe the Agent can operate this transformation.

Mental Spaces are general patterns set up and invoked while we talk, listen, write and read. Maybe a mental space may be represented/metaphorised with a workspace of all the elements called or invoked by the user.

Blendings refers to a mental space configuration in which elements of two input spaces are projected into a third space, the blend, which thus contains elements of both but is distinct from either one.

The language of thought hypothesis claims that meanings can be described by combining semantic primitives. These primitives are universal.

Some of many open questions:
1. Are image schemas scale–free?
2. What concepts are amenable to grounding?

Two interesting links this w.e.. Both relates to my idea of increasing the connectedness of strangers through providing a network of help or a network of resources. The first one is called Pocket Rendezvous and is an implementation of the Apple Technology for the PocketPC. There is also an interesting link to a Rendezvous implementation for Windows and Linux. [more…]

The second link is called Crowdsurfer which enable the user to surf for other Bluethooth devices in the vicinity and get in contact with them. It is primarily designed for a campus environment.