Posted: Sat Jul 08, 2006 3:09 pm
So I'm doing a project for a biz professor. He needs some databases merged. For the past four months, I've been trying to get access to the resources to crunch numbers for him. In the meantime, I've researched alternative methods so I can make sure I'm maximizing correctness. The problem is that the methods which seem to give the best results are written by the people w\ the worst grasp of how to write a paper explaining shit.
For instance, the paper I'm currently reading talks about "distance" and "similarity" functions that tell you something about two strings. The closer two strings are, the higher their similarity, but the smaller their distance becomes. Despite meaning the precise opposite shit, the paper uses these terms interchangably.
The psychosis doesn't end there. Part of the score for whatever function they're actually referring to is based upon frequency lists of words in their respective documents. Imagine I've got a word in a phrase from document A & I'm scoring it. It requires me to look up the frequency of the word in document B (amongst other shit). If the word isn't in document B, it causes a divide by zero error. Either they're idiots or they can't express themselves. There's numerous little pitfalls like this.
Christ help me cos this is one of the more well written papers I've come across. Can't academics just start putting ghostwriters in their budgets from now on so I can understand what the fuck they're saying?
Edited By Malcolm on 1152385908
For instance, the paper I'm currently reading talks about "distance" and "similarity" functions that tell you something about two strings. The closer two strings are, the higher their similarity, but the smaller their distance becomes. Despite meaning the precise opposite shit, the paper uses these terms interchangably.
The psychosis doesn't end there. Part of the score for whatever function they're actually referring to is based upon frequency lists of words in their respective documents. Imagine I've got a word in a phrase from document A & I'm scoring it. It requires me to look up the frequency of the word in document B (amongst other shit). If the word isn't in document B, it causes a divide by zero error. Either they're idiots or they can't express themselves. There's numerous little pitfalls like this.
Christ help me cos this is one of the more well written papers I've come across. Can't academics just start putting ghostwriters in their budgets from now on so I can understand what the fuck they're saying?
Edited By Malcolm on 1152385908