Requirements
The main requirements that a clustering algorithm should satisfy are:
scalability;
dealing with different types of attributes;
discovering clusters with arbitrary shape;
minimal requirements for domain knowledge to determine input parameters;
ability to deal with noise and outliers;
insensitivity to order of input records;
high dimensionality;
interpretability and usability.
MSTROHM: "Why lists won't become superfluous."
The list is the origin of culture. It's part of the history of art and literature. What does culture want? To make infinity comprehensible. It also wants to create order -- not always, but often. And how, as a human being, does one face infinity? How does one attempt to grasp the incomprehensible? Through lists, through catalogs, through collections in museums and through encyclopedias and dictionaries.
[...]
In the case of Google, both things do converge. Google makes a list, but the minute I look at my Google-generated list, it has already changed. These lists can be dangerous -- not for old people like me, who have acquired their knowledge in another way, but for young people, for whom Google is a tragedy.
Last week, Diaspora — the OSS privacy-respecting social network — released a “pre-alpha developer preview” of their source code. I took a look out it, mostly out of curiosity, and was struck by numerous severe security errors. I then spent the next day digging through their code locally and trying to get in touch with the team to address them, privately. In the course of this, I mentioned obliquely that the errors existed on Hacker News, and subsequently was interviewed by The Register and got quoted in a couple of hundred places.
Once you’re through with this article, there’s a followup that shows how to also make them clickable. Many people find the object-oriented aspect of JavaScript to be very confusing. Fortunately, the designers of the Google Maps API have managed to ma
"Here's a preliminary data mining analysis of musical social networking service Last.fm. An automated classification into clusters or sub populations with related musical genres reveals the structure of musical preferences among the users in a relatively large sample population. Musical tag clouds are adopted to characterise users and populations, which adds a highly descriptive value and aids with the interpretation of the results."
PPrett: The major clou is that the topical clusters (e.g. Information Retrieval, Machine Learning) can be sorted according to their HITS scores (Hub, Authority).
accessing_zotero_data_using_an_sqlite_client Whenever possible, people writing utilities that access Zotero data should use the Zotero data API from within Firefox, as it provides both a consistent interface to the data regardless of the underlying str
Introduction This is the manual for the Sage mathematical software system. Sage is free open source math software that supports research and teaching in algebra, geometry, number theory, cryptography, and related areas. Both the Sage development model an