Monthly Archive for September, 2005

Designing ecosystems

Published on 9/29/2005 in Diary. Closed

I found this last post of Martin Terre Blanche extremely intriguing.

George Siemens is definitely onto something with his ideas about networked learning. His latest contribution is a piece arguing that learning design should be about designing ecosystems rather than about designing courses:

“My latest contribution is a piece arguing that learning design should be about designing ecosystems rather than about designing courses:

“We should be focusing on designing ecologies in which learners can forage for knowledge, information, and derive meaning. What’s the difference between a course and an ecology? A course, as mentioned is static – a frozen representation of knowledge at a certain time. An ecology is dynamic, rich, and continually evolving. The entire system reacts to changes – internal or external. An ecology gives the learner control – allowing her to acquire and explore areas based on self-selected objectives. The designer of the ecology may still include learning objectives, but they will be implicit rather than explicit.”

I definitely subscribe to this idea of rethinking the way we propose courses and we think about teaching. Personally I strive for a middle-earth approach between the open ended, multi-brainset, constructivist approach and the self-regulated, computer assisted collaborative learning. I think the way in should be finding a script that structures the interaction and leaving enough room in that script to support multiple attitudes/abilities.

Tags: , ,

FieldTools: Collect, Organize, and Share Your Biology Research

Published on 9/29/2005 in Diary and Links. Closed

A set of tools for field biologists to collect, organize, and share their research. These tools comprise mobile components (e.g., an in-the-field photo annotation tool) and lab components (e.g., the ButterflyNet interface for viewing multiple data sources linked by their metadata, such as time and location).

Fieldtools

Tags: , , , , , , ,

Graffiti Archeology

Published on 9/29/2005 in Diary and Links. Closed

Graffiti Archaeology is the study of graffiti-covered walls as they change over time.  The grafarc.org project is a timelapse collage, made of photos of graffiti taken at the same

location by many different photographers over a span of several years. Most of the photos are from San Francisco, over a timespan from the late 1990’s to the present.

Using the grafarc explorer, the user can visit some classic graffiti spots, see what they looked like in the past, and explore how they have changed over the years. One of the part I like the most is the timeline bar. Something that reminds me so much of the timeline I developed for the Biosphera project. This one is nicer though.

Graffiti-Arch-2

Tags: , , , , , ,

coding like a pig: a Part of Speech Tagger module for STAMPS

Published on 9/28/2005 in Diary. 1 Comment

Now I understand why sometime Jamie or Fabien are not blogging much.  In fact I spent a couple of days on my Part of Speech Tagger, that should extract relevant keywords from the geolocalised messages left by the users in the STAMPS system.

After a couple of hacks around the POSTagger of the University of Stuttgart, I managed to write my own Python extension (thanks to Shuja and Patrick for the hints).

It doesn’t output much at the moment, or at least the output is not so self evident. However I added some lines in the logger to have some sense of stats:

2005-09-28 18:34:49,625 – main – INFO – — Tagging session started

2005-09-28 18:44:26,705 – main – INFO – The number of messages tagged is: 190

2005-09-28 18:44:26,710 – main – INFO – The number of new tags created is: 282

2005-09-28 18:44:26,710 – main – INFO – The number of messages dumped is: 3

2005-09-28 18:44:26,713 – main – INFO – — Tagging session ended

Tags: , , ,

How to handle a Many-to-Many relationship with MySQL

Published on 9/24/2005 in Tech. Closed

Shuja pointed me to this article that explains how to set many to many relationships within MySQL. The trick is to use a cross references table that will divide the many to one and one to many link.

Many-To-Many-03

Tags: , ,

Aquamacs

Published on 9/21/2005 in Tech. Closed

Aquamacs

Aquamacs is an Aqua-native build of the powerful Emacs text editor. By “Aqua-native,” we mean more than just the fact that this version of Emacs runs as a standard OS X application.  Aquamacs features extensive customization that enables it to conform better with Apple’s standard Human Interface Guidelines (HIG) than standard versions of the editor do.

Aquamacs-Screenshot.1

Tags:

Spatial Clustering Methods in Data Mining: A Survey

Published on 9/21/2005 in References. 23 Comments

J. Han, M. Kamber, and A. K. H. Tung. Geographic Data Mining and Knowledge Discovery, chapter Spatial Clustering Methods in Data Mining: A Survey, pages 1–29. Taylor and Francis, 2001. [url]

—————————

This paper presents a review of spatial clustering methods, which are considered an important component of spatial data mining. The authors classify the methods into four categories: partitioning method, hierarchical method, density based method, and grid-based method.

Partitioning methods like the k-means , the k-medoids and EM clustering are methods which make uses of a techinque called iterative reallocation to improve the clustering quality from an initial solution. These methods tend to find clusters that are of sperical shape and they are made for minimising the disctance from the data objects to their distance centers.

On the contrary of these, hierarchical clustering algorithms fixed the membership of a data object once it has been allocated to a cluster. BIRCH, CURE and CHAMELEON uses complex criteria for compressing and relocating data before merging clusters.

A third group of these methods is based on density of data points within a region to discover clusters. Belong to this category methods like DBSCAN, OPTICS and DENCLUE.

Finally, to increase the efficiency of clustering, grid-based clustering methods approximate the dense regions of the clustering space by quantizing it into a finite number of cells that contain more than a number of points as dense. Clusters are then formed by connecting the dense cells. To this category belongs STING, and CLIQUE.

Tags: , , ,

Some reflections towards the clustering algorithm

Published on 9/20/2005 in Diary. Closed

1. Latent Semantic Analysis is a technique that is possible to use to define the similarity between the messages. So far I could not find a good implementation in Python, except the one by James Stanley. The pros: it is reliable and based on solid statistical methods. The cons are that it works with a fixed corpus that is given by the authors and that does not change over time (supervised method).

There are other approaches still pointing on google or other web resources. See for instance this paper by Marco Baroni. Using LSA we can ultimately graph something like this:

200509201602

2. Why clustering? What we want is to support the user in the exploration of space. The cluster then reflect to the concept of a Landmark, a pinpoint of “features” to the physical space.

3. About LSA. Once computed the similarity with the LSA algorithm, we then have only to cluster points with numerical feature using a geographical criteria. We loose the semantic dimension. We computer the LSA similarity only on the keywords set to avoid noise.

4. The clusters are static within a certain period of time. Periodically they are updated. A search query can be mapped against the clusters’ keywords (a composition of the message keywords that have generated them) using again LSA this will give a gradient of matching results within the DB content. A simplification might be:

200509201612

5. The cluster machine. What is still missing here is the inner working of the cluster machine. In fact, I have to clarify wether this will be based on an existing method or derived by one of this into a custom made formula. I am currently reviewing a survey of these methods. Where I discovered the Non-Spatial-Data-Dominant Generalization that should provide more flexibility in the aggregation of data.

Tags: , , , , ,

Questsin: a google set clone

Published on 9/20/2005 in Diary and Links. Closed

I found this google set clone that seems to be a bit more reliable and has some nice features, like an improvement of the search results. It was done by Nicholas Manolakos. He parsed over 100 million tables so far and created google sets functionality.

1. minus sets no just add them.

2. sort based on average distances.

3. reverse google sets and get the parents, for example “dog”,”cat”,”mouse” might return pets and “pig”,”cow” farm animals.

A very interesting piece of work. Thanks Nicholas.

Questsin

Tags: ,

Google Clustering

Published on 9/20/2005 in Diary. Closed

It seems that Google is playing with clustering. Some of their server group the results into functional blocks.

Google-Clustering-Small

(via)

Tags: ,

Surface Patterns

Published on 9/20/2005 in Diary and Links. Closed

Surface Patterns: Audio Tours uses a Global positioning System [GPS] device to explore how memory is linked to urban and domestic place. The GPS device can only describe latitude, longitude and altitude; however, when used to trace the route that someone takes through a place, it can reveal the pattern of the path taken, allowing us to share knowledge of hidden locations and unexpected vantage points along that path.


Traditional maps tell us where landmarks are, what streets are called, and where to find the centre of town, whereas the subjective histories and stories ezplored in this work are played out over time and rely on very different ‘memory maps’.


The installation uses contributions of unwanted wallpaper, pasted on the gallery walls and threaded or punctuated with GPS patterns of ten walks. Traces of audio recordings made in conversation with the walkers are played back through speakers embedded in the walls behind the wallpaper.

Inviteandfront

Tags: , ,

Learning with Tangibles

Published on 9/19/2005 in Diary and Links. Closed

I have been following a couple of sparse threads on this subject and I thought that was worth sharing some pointers on this matter. The first one was one post on wmmna about “Smooth Shape manipulations“. This reminded me of the brilliant literature review on the subject by Claire O’Malley founded by Nesta Future Lab. Then Nicolas has his own weekly review of tangible interfaces on his blog, and recently he pointed on this special issue of International Journal of Design Computing.

I have been always fascinated by these kinds of approach to learning technologies since I started working at MLE and I had a chance to play with a prototype of DinoStable.

Tags: ,

Sardinia mon amour

Published on 9/19/2005 in Personal-Fun-Unrelated. 2 Comments

Back from holidays … sigh! Here are some of the pictures I took that will light my hearth during this cold winter. Besides the place is Stintino in the province of Sassari in the north-west corner of the island. Enjoy.

Img 1250

Img 1252

Img 1299

Img 1308

Img 1300