Saturday, April 30, 2005

I got a OS X Panther for Dummies, Fedora Core Linux for Dummies and an O'Reilly book on the Palm Pilot all for 5$ at a computer book store closing sale. I might go back there later today get some more books. I also want to look at microwaves later today. I also have to take the paper version of my final project to school today.
I flipped through Sherly Hamilton's MA and then found a reference to Susan Hacker who Dawn Moore had meantioned. I then surfed the library web site and found Hacker's book. I clicked on some links in her books library listing or as we used to say index card. I then surfed the MA collection on-line and found some MA's on feminism. Then looked up a fellow workers MA and then some MA's that my honours paper supervisor had been an advisor for. I then looked up M.Sc.s that Shirley Mill's has been an advisor for. Then I ordered a download of all these and the download instructions are being sent to my school email address.

Friday, April 29, 2005

I send my final project files to the professors Dr. Mills and Dr. Normington. I may not work on this anymore. It cost me my attendence at the symposium in crime prevention and disablities but it was fun and challenging to do this final project. I also learned quite a bit about R the open source statistical software.

Maybe this data mining course will allow me to attend graduate school in statistics. I am really on holidays from school now. I will start to get ready for my public law course this summer.

Thursday, April 28, 2005

I stayed up Tuesday until about 4 PM Wednesday. I volunteered to staff a union voting table during Wednesday late morning and early afternoon. I had a good talk with a PhD. student in computer science about data mining and education for mature students. I also chatted with our union's business agent about the new deal our federal government made with the New Democrats to keep the government alive. Ya ya I was pretty chatty after no sleep. When I got home I blew my top and there is no excuse for that.

I read my lecture notes from my data mining course and then today was able to get a trees classification to at least plot although comparing residuals did not work for a basic trees proceedure and when I did a confusion matrix for an rpart proceedure it looked pretty bad. Again I am favouring logit regression as the best result seems to be with logit regression. I did work out a test of logit regression but don't seem to have a print out for it yet. The test was to build the logit model with the test data using the model from the training data but I made all kinds of errors trying to do that at the different dimensions I had used for logit. I got that done today. I worked for about five hours today on the final project.

Tuesday, April 26, 2005

I used yahoo.ca's video search to search out video's about statistics. I found some basic statistics videos. I will try to watch them on my PSP. I also searched out Green Party videos.
Well I divised my own code finally to test a model using the found model's predictive abilities on the test data. I spent about an hour and a half on this so I did work on homework tonight. So I will now write up the logistic regression models and move on to other models. I have at least four more techniques/models to try before I can say I am done.

I will keep working on this assignment/project which is due Saturday and only if I get it done will I be able to attend the symposium at the end of this week. I have about 35 hours until the symposium begins.

Since I am going to a Crime Prevention and Disabilities symposium this week I searched Lexis and the library's journals and found some articles but nothing specifically right on the money. I guess this is a new field.

I may not be able to attend after all with all my ebay surfing I still have a lot of statistics studies to complete before April is over. I also want to volunteer for the contract instructors union because they are holding a ratification vote this week.

So I may stay up late and work on statistics homework and then go into the university in the morning and end my day staffing the vote table as a volunteer. I would read at the table and then come home and sleep at 4 PM Tuesday.

I tried some full model logistics regressions and some pared down versions. I also tried the covariance greater than 0.09999 and that didn't look good. So the best logit model I found was for just the first three variables. I spent an hour on this tonight. So I have spent about 14 hours on this final project so far. I need to spend another 76 hours still. I am just printing my R session form this work.

Monday, April 25, 2005

I did not stick to my planned hours on the data mining final project. I have only put in one hour today and I planned to put in 11.

I did a logistic regression of all 24 variables. Using a book on how to interpret logistic regression in R I have now to pare down my model. My next step is too use the covariance cut off of 0.1 and try that model.

Saturday, April 23, 2005

I have not worked on data mining yet tonight. I did do some web work. I also talked about technology and cell phones with my brother.
After some tweaking of my code I was able to run a glm command and get a printout. I will take a little break now and then review the logistic regression interpretation and see what I have got. Well at this point I may have at least one classification model so it was worth my while to get down to working on this. I have worked for 3 hours now out of 48 so 45 left to go. I have one more hour to do before doing other unrelated work.
I am running my first logistic regression and must wait for the computer to finish printing the plots.
I just spent an hour thinking, reading lecture notes on logistic regression and am about to try to do a logistic regression for the first three variables. My work computer is a bit slowed because I am still printing the covariance plot. I will end up with four copies of this plot. I spent the last ten minutes making an open office document of my code and my intent with this is to have a good copy of my code in a word processor. Of course I also need a copy documenting everything I tried and my mistakes.
I actually can read the covariance scores on the plot. I just looked at it and can identify the high scores for relation with the class variable. At this point I may pick a cut off value of say .0999 and take all the variables relating better than this and try a next step with this. These would be at´s 1,2,3,4,5,6,9,11,17,21. This is a rough covariance tree to decide which variables look promising for use in the model. Using a covariance score cut off of .199 I would use at´s 1,2,3 only. This makes sense of the data which also showed structure when looking at the first at´s in ggobi. So I think this is progress.
I did one hour of my data mining project. I started xess and then used my code from yesterday and copied some code from the first lecture to produce a plot of relationships between the 25 variables with a matrix style plot. The lower triangle shows the xy relationships. The diagonal shows the histograms of each variable. The upper triangle shows the covariance score. Although I could do this and am now getting the plot printed four times, the plot is showing too many variables to see the plots in other words the plots are too small. We were warned about this.

I took a break and then wrote this journal entry and now have three hours before I do a different job and 47 hours left until the project is completed in terms of time put in.

Friday, April 22, 2005

I am not working hard on my data mining assignment. I should put in about 88 hours on this assignment. I have put in about 8 hours now. I guess the thing to do is plan out the other 80 hours. I can´t work on it on Sunday or Thursday or Friday or next Saturday, because of work and a symposium the end of next week. I can put in 12 hours today and sleep at 2 PM. This would give me maybe two hours to work on the newspaper route. Then Monday I could put in another 12 then Tuesday another 12 or so. Then Wednesday I could put in a final 12. So I could maybe put in about half of what is required. That´s a plan then I will start tonight by reviewing all the notes on supervised learning by 4 am. That gives me about an hour and a half per lecture. I´ll get started now.

Thursday, April 21, 2005


This is a view of my workplace. The School of Mathematics and Statistics at Carleton University. Posted by Hello
I got my grade yesterday for my honours paper and it was good and what I hoped for given my weak conclusion and weak development of my hypothesis.

I am a week and a half through the readings for my summer public law course. The course starts May 18th.

I applied to be a teaching assistant again next term. I will include a picture of our school of mathematics and statistics here.

Tuesday, April 19, 2005

I am reading more about industrial safety in schools. I spent an evening having a volunteer appreciation dinner with city officials and the Chief of Police. This is for my victims support volunteering. I won some chocolates in the raffle we had.

My teaching assistant work is now complete for the winter 2005. I have applied again to work in the summer.

I have a data mining project still to complete. Then this summer I am studying public law. I have the public law textbooks purchased and have been reading them in preparation for this summer.

Sunday, April 17, 2005

I am just completing the last of my marking duties for my statistics TA work today. I will return the graded assignments tomorrow to the professor Ann Woodside. I believe the students write their exam on Tuesday.
I am reading Hamilton, Sheryl N. Intimate Couplings: A Feminist Interrogation of the Cyborg (M.A. Thesis, Carleton University, School of Journalism and Communications, 1995) [unpublished].

Wednesday, April 13, 2005

I am busy marking and reading today.
I still have some grading to do from the winter term. I hope to complete one of these grading tasks today.
I have been read about industrial safety this morning, yesterday and Monday. I have also been reading about feminist science. This morning I also read about GLM's.

Tuesday, April 12, 2005

My honours fourth year paper is completed and I handed it in to my supervisor yesterday. I have read a little but mostly relaxed for the last 24 hours.

Sunday, April 10, 2005

The Bruce Sterling footnotes are now done but I added another source to get the sentence of Morris because Sterling doesn't really cover Morris. The other source is
Cavazos, Edward A. & Morin, Gavino. Cyberspace and the Law: Your Rights and Duties in the On-Line World (Cambridge, Mass.: MIT, 1994).
This was the very first book I owned on computers and the law a good old stand by source that has proved its usefulness again.
I have three cases left to brief, a few footnotes to make to Bruce Sterling's The Hacker Crackdown. Then I just need to make correct footnotes and bibliographic entries for Statutes and I am done. Then I can read it through tomorrow and print it to hand in a 2 PM. I might stay up another 4 hours tonight. The Internet is working my printer is working. Things are alright for getting this law paper done.
I have 28 pages pretty much done except for a few footnotes. I have another 15 or so pages to complete. This only amounts to a few pages of new writing. Then I need a final footnotes check and I am done. The end is near.
I am up over night putting the final touches on my laws honours paper. It is looking pretty good now and is due tomorrow.

I then only need to work and study on statistics until May. Of course, the break in my own classes will allow me to read more widely and catch up on web mastering work.

Sunday, April 03, 2005

I started to restructure my paper. I am just going to take a break now. My lecture notes for Monday´s data mining class are all printed out now with no printing errors.
I am going to have to give up on my own school work today at least my data mining course. I will continue to print the notes for tomorrow´s lecture but right now for the next two hours I am going to brief cases.
I finally got the separate histograms printed for the spam emails and the non-spam. I was able to see from these printed side by side that classification trees will probably work well here. I am having trouble with the gini tree program but am going to ask Ted Normington about this error with the min function in my code. I am also printing out tomorrow´s lecture notes and hope to get these notes printed tonight before I sleep.
Here are my present school work assignments:
  1. Mark the STAT2507C4 assignment #5. Due Wednesday April 6th.
  2. Mark the STAT3507 assignment #3. Start this again on Tuesday April 5th. Due some time later in the week.
  3. Mark the STAT3507 assignment #4. I can´t start this until next week. Due April 16th or so.
  4. Complete assignment # 3 for my STAT4601 course in data mining. This assignment involves making a spam filter using supervised learning. Due Monday April 4th.
  5. Complete my honours paper for LAWS4908. I need to brief cases in my own words and rewrite my whole paper. Due Monday April 11th.
  6. Complete my STAT4601 final project. This project is also supervised learning based on real German credit data. The categorization should determine good and bad credit risks. Due the end of April.
The pope has died. I grieve with the rest of the world a champion of peace and tolerance. I am sorry that he has died. I am not that religious but agreed with many of the popes statements on peace and understanding.

Friday, April 01, 2005

I have a data mining assignment to do as I upgrade my statistics education. The assignment is to filter spam. I have been given some 4601 observations or emails. I have 57 variables based on word frequency, character frequency and capitals frequency and length. Then the 58th variable is 1 or 0 depending on whether the email is spam or not. So this is supervised learning. I was able to open this data set in both R and SAS. In SAS I did a proc means and also a proc chart. I have 58 histograms coming out of SAS. I am printing these histograms 4 per page on both sides of the page. I am also printing the results of the proc means.

I am still reading Tomasina Borkman on self help groups and experiential knowledge and have been reading the chapter on professionals over night. I have also started to read abook on African telecommunication development. I also read more about educational reform in the USA in the 1990´s from a book on Urban eco-systems and teaching urban eco-systems. Last night I read a little bit more about a feminist Supreme court Justice.