Jon – Page 7 – uClassify blog

Stock prediction results week 20

In short, a much longer evaluation period is needed to verify if there is any significance in those classifications. It did quite well for Tuesday and Wednesday (60% and 83% correct), bad on Thursday (36%) and what can be expected by random on Friday (48%). I may setup a site that will feed predictions automatically over a long period of time. What also would be really interesting is to feed stock news into the training data. Below is detailed information of how the classifications turned out, if anyone is interested!

Friday 15/5

Predicted winners, actual winners:TXI, AVP, AAWW, DD, GEF, WCC, CTSH, ACTG, HIT, AKS, DRQ, AVT, ALB, ACXM, BWS, LTD, CBT, CHRS, DWSN, DUG, FXP, BMS, GYMB, PETM, AFFX, COL, ALOG, ADVNB, AXE, AN, WIT, MDC, DY, OIS, JOSB, IBI, ASPM, TEX

Predicted winners, actual losers: GPI, ANAD, FWRD, KMT, ATPG, JBLU, ESL, USG, ALV, KBR, ANF, AYR, STAA, ADPT, TSO, LINTA, ME, GIB, BC, AIR, BGC, USTR, CBI, EXPE, COG, FCG, ATI, EAC, DOW, ETH, AYI, CIM, SCHN, FNM, ARW, TER, TKTM, AMSG, CEDC, AUO, TRW, ENER, GLBL, CHL, CONN, OI, JOE, HXL, PLL, ADBE, CHU, EMN, COH, APC, CPX, GGB, EEQ, TIE, ENG, AFL, A, FO, X, JRCC, AAI, KBW, CHRD, CAJ, CTX

Predicted losers, actual losers:UNT, GT, ABT, WAB, AWI, WGOV, EQIX, CLR, CRBC, CAM, UEPS, CLF, IP, DEI, UYM, DVN, LDSH, WRES, UBS, UFPI, GLF, WHR, DVR, EGN, NOA, CHK, JCI, CEO, CBR, FIW, DSI, ACV, XEC, HTCH, ENZN, OC, AMR, CRS, KWK, WIN

Predicted losers, actual winners:TTMI, UTSI, TWI, ID, DCI, CNTF, GNK, AAPL, WWW, GGG, COMS, UA, GWR, ARQL

Accuracy (78/161) : 48%

Thursday 14/5

Predicted winners, actual winners:ESL, MDC, PLL, PETM, DVN, COG, LDSH, ACV, TSO, CRS, DLB, PTNR

Predicted winners, actual losers:FXP, DUG, ADPT

Predicted losers, actual losers:KMT, WAB, CRBC, DWSN, ADVNB

Predicted losers, actual winners:CPX, TTMI, UNT, FNM, HIT, UTSI, GT, AWI, GIB, GPI, DD, ME, ANAD, CHRD, TXI, WGOV, EQIX, ATPG, DOW, AIR, ALV, AVP, USG, CLR, ACTG, ATI, CIM, ACXM, TWI, CAM, ID, UEPS, COH, AYR, CLF, BWS, ALOG, CEDC, ANF, IP, DCI, GNK, UYM, AFL

Accuracy (17/47) : 36%

Wednesday 13/5

Predicted winners, actual winners:

Predicted winners, actual losers: HIT, EGN, GIB, DLB, TTMI, GT

Predicted losers, actual losers: FWRD, DWSN, CEDC, UTSI, FNM, KMT, AWI, TXI, ANAD, ME, WGOV, GPI, ATPG, AVP, EQIX, DOW, AIR, ALV, CLR, USG, CRBC, ATI, ACTG, CIM, ACXM, ID, TWI, CAM, AYR, CLF, BWS, USTR, FCG, EAC, DEI, WCC, EXPE, CBI, ALB, GLF, UFPI, WHR, AVT

Predicted losers, actual winners: ESL, ADPT, ANF

Accuracy (43/52) : 83%

Tuesday 12/2

Predicted winners, actual winners: DWSN, UFPI, FXP, ESL, DUG, EXPE, DLB, CONN, DD, COMS

Predicted winners, actual losers: CEDC, DCI, AVT, FWRD, CTSH, GEF, ALV, AUO, DRQ, UNT, AKS, XEC, TXI, FCG, EGN, BWS, AAPL, GIB, COG, AWI, DOW, WCC, EAC, CIM, FIW, EQIX, HIT, USG, ETH, AYR, UEPS, AAWW

Predicted losers, actual losers: WWW, ATI, CP, CRBC, ECA, WHR, TRW, DISCA, GGG, USTR, GWR, CBI, DISH, CLF, ALB, CRS, AYI, AMSG, FFIV, CHL, ARW, DVR, TER, CHRS, GLS, CBT, ACXM, GLBL, UBS, DSI, GT, TTMI, CVC, ANF, TWI, FNM, CKP, ADPT, TSO, CLR, GPI, ANAD, AVP, CAM, GLF, ATPG, CBB, ACTG

Predicted losers, actual winners: GLT, TKTM, ENER, DEI, CBR, UTSI, AIR

Accuracy (58/97) : 60%

Attempt on predicting stock for 15 of May 2009

Finally friday! Todays classifier predictions seems to vary more than yesterdays.

Prediction for 15 of May 2009

It’s now the 15 of May 08:15 in the morning (GMT+1, Stockholm), and this is a prediction for today:

Some predicted winners (closing price > opening price):
GPI, ANAD, TXI, FWRD, KMT, ATPG, AVP, JBLU, AAWW, DD, ESL, USG, ALV, KBR, ANF, AYR, GEF, STAA, ADPT, TSO, LINTA, ME, GIB, WCC, BC, CTSH, ACTG, AIR, BGC, HIT, AKS, USTR, CBI, DRQ, AVT, EXPE, COG, FCG, ALB, ATI, ACXM, EAC, DOW, BWS, ETH, LTD, CBT, AYI, CHRS, CIM, SCHN, FNM, DWSN, ARW, TER, TKTM, AMSG, DUG, CEDC, FXP, AUO, TRW, ENER, GLBL, CHL, CONN, OI, BMS, JOE, GYMB, HXL, PETM, PLL, ADBE, CHU, AFFX, COL, ALOG, EMN, COH, APC, CPX, GGB, ADVNB, EEQ, AXE, TIE, ENG, AFL, AN, A, FO, X, WIT, JRCC, MDC, DY, AAI, KBW, CHRD, OIS, JOSB, IBI, ASPM, CAJ, CTX, TEX

Some predicted losers (closing price < opening price):
TTMI, UNT, UTSI, GT, ABT, WAB, AWI, WGOV, EQIX, CLR, CRBC, TWI, ID, CAM, UEPS, CLF, DCI, IP, DEI, UYM, CNTF, DVN, LDSH, GNK, WRES, UBS, UFPI, GLF, WHR, AAPL, DVR, EGN, NOA, CHK, JCI, CEO, WWW, CBR, FIW, GGG, DSI, ACV, XEC, COMS, UA, HTCH, ENZN, GWR, OC, AMR, CRS, ARQL, KWK, WIN

Attempt on predicting stock for 14 of May 2009

It turns out that all this manual work takes a lot of time, so I’ll automate the process during this weekend. However, I will continue to post predictions. As soon it has been automated I will post accuracies, which is going to be interesting =)

Again today, it was hard to find predictions for stock that may go up, out of the 200 most reliable classifiers, only 15 were predicted as winners.

Prediction for 14 of May 2009

It’s now the 14 of May 08:30 in the morning (GMT+1, Stockholm), and this is a prediction for today:

Some predicted winners (closing price > opening price):
ESL, FXP, MDC, DUG, PLL, PETM, ADPT, DVN, COG, LDSH, ACV, TSO, CRS, DLB, PTNR

Some predicted losers (closing price < opening price):
CPX, TTMI, UNT, FNM, HIT, UTSI, KMT, GT, WAB, AWI, GIB, GPI, DD, ME, ANAD, CHRD, TXI, WGOV, EQIX, ATPG, DOW, AIR, ALV, AVP, USG, CLR, CRBC, ACTG, ATI, DWSN, CIM, ACXM, TWI, CAM, ID, UEPS, COH, AYR, CLF, BWS, ALOG, CEDC, ADVNB, ANF, IP, DCI, GNK, UYM, AFL

Attempt on predicting stock for 13 of May 2009

I am in a hurry and didn’t have time to predict all winners, I’ll have to automate this =) I will get back with the results from yesterday later.

Prediction for 13 of May 2009

It’s now the 13 of May 08:30 in the morning (GMT+1, Stockholm), and this is a prediction for today:

Some predicted winners (closing price > opening price), incomplete list:
HIT, FWRD, EGN, GIB, DWSN, CEDC, DLB

Some predicted losers (closing price < opening price):
TTMI, GT, UTSI, FNM, ESL, KMT, AWI, TXI, ANAD, ME, WGOV, GPI, ATPG, AVP, EQIX, DOW, AIR, ALV, CLR, USG, CRBC, ATI, ACTG, CIM, ADPT, ACXM, ID, TWI, CAM, AYR, CLF, BWS, ANF, USTR, FCG, EAC, DEI, WCC, EXPE, CBI, ALB, GLF, UFPI, WHR, AVT

Tomorrow I will report how many that were correctly classified, until then you may check for yourself at for example Yahoo! Finance.

Attempt on predicting stock for 12 of May 2009

A couple of months ago I wrote about a project that would use classifiers to predict tomorrows stock market. I’ve now found some time to implement the very basics, a database with stock history, +3000 classifiers trained on this data and some simple classification scripts.

Instead of waiting, pondering and validating that everything works properly I’ve decided to do this “live”, starting to predict what may happen with some of the stocks for today. There are probably some corrections that will need to be done on the way to see if this is working or not (bug fixes etc).

I’ll select stocks that seem to have been working well to for classification in the past, I’ll probably present 100 of them per time.

Prediction for 12 of May 2009

It’s now the 12 of May 08:00 in the morning (GMT+1, Stockholm), and this is a prediction for today:

Some predicted winners (closing price > opening price):
DWSN, CEDC, DCI, AVT, FWRD, CTSH, UFPI, GEF, ALV, FXP, ESL, DUG, AUO, DRQ, EXPE, DLB, UNT, AKS, XEC, TXI, CONN, FCG, EGN, BWS, AAPL, GIB, COG, AWI, DOW, WCC, EAC, DD, CIM, FIW, EQIX, HIT, USG, ETH, AYR, UEPS, AAWW, COMS

Some predicted losers (closing price < opening price):
WWW, ATI, CP, CRBC, ECA, WHR, TRW, DISCA, GGG, USTR, GWR, CBI, DISH, CLF, ALB, CRS, GLT, AYI, TKTM, AMSG, FFIV, CHL, ARW, DVR, ENER, TER, DEI, CHRS, GLS, CBT, ACXM, GLBL, UBS, CBR, DSI, GT, TTMI, CVC, ANF, TWI, FNM, CKP, ADPT, TSO, CLR, UTSI, GPI, ANAD, AVP, CAM, GLF, ATPG, CBB, ACTG, AIR

Tomorrow I will report how many that were correctly classified, until then you may check for yourself at for example Yahoo! Finance.

uClassify in the press

We’ve had hundred of thousands of mentions throughout the blogosphere and I’m really thankful for this!!! I’ll try to update this post as we go along. Please comment if you can help me with this list!

Here are some I remember/managed to find:

CNBC blogs about Mattias Östmars genial invention Typealyzer, 2008-03-25.

Business Week writes about Typealyzer, 2009-03-22.

Norweigan newspaper test author recognition with uClassify, 2009-03-04.

Article about uClassify (only in paper version), 2009-01-01. Germans really seem to be interested in this kind of stuff!

Article about uClassify, 2008-12-07.

Kevin Kelly mentions Typealyzer, 2008-12-05.

Doc Searls mentions Typealyzer, 2008-11-30.

German Daily Taz
Another Genderanalyzer interview, 2008-11-12.

Germanys biggest daily newspaper with a circulation of 450 000 copies. Genderanalyzer interview, 2008-11-10.

Genderanalyzer is featured 2008-11-03.

Trve or Emo?

Another interesting uClassify web application has seen the light, Trve VS Emo. It tests a site if it’s “True Black Metal” or “Emotional”. The author, Albert Örwall writes

To “train” the trve-classification I’ve used lyrics by norweigan black metal bands, such as Mayhem, Burzum and Darkthrone. The emo-classification is based on lyrics by emo bands like My Chemical Romance and Fall out boy…

I tested with a hard rock blog I found randomly, Hard Rock Hideout which proved to be 81% Trve (true black metal). I then tested with this blog which turned out to be 100% Emo 🙂

Is there any need for automatic music tagging?

This is really cool, another cool thing would be a classifier that has been trained on texts from all genres (hip-hop, country, soul etc), this would not only be a fun way to test your blog it could also be used for automatic lyric tagging (hence track and album tagging). Does anyone know if there is any need for such a web service?

An experiment – predicting the stock market

During my three weeks of vaccation, I had an interesting conversation with a company that has a bot that trades soccer bets. This inspired me to set up classifier model that tries to predict the stock market.

Seeing into the future by looking on the past

My idea is to start with something that is really simple to implement and test – predicting the stock market tomorrow based on the history. I have no particular interest in stocks and the only thing I really know is that you should buy cheaper than you sell, hence this is the only thing the classifiers know as well =)

What’s going to happen tomorrow?

Setup

For training data I’ve used historical stock prices that are downloaded from Yahoo Finance. Then I’ve automatically created one classifier per stock (in total about 3100 classifiers) that, given the todays stock state predicts tomorrows by inferring over historical data. From the 3100 classifiers (stocks) I will pick the top X classifiers that are most confident. This is done by evaluating the training data and picking those that historically would have worked best. This is the most time consuming task takes several hours to run 10-fold-cross-validation.

Evaluation

As I evaluate this project I will post predictions in this blog and follow up with the accuracy when the correct answers are known.

More to come!!

What’s going on?

The last three weeks I’ve been doing some other things than working on uClassify.com. I really needed a break to regenerate some energy and inspiration. During this period I’ve been talking to Roger Karlsson about what we should develop next on uClassify.

uClassify HTML forms

We know that there are a lot of cool classifiers being built but not many of them finds their way to the public in a web application. We believe that the reason is the last barrier – using the API and to do that you need to write code. Our plan is to allow users who doesn’t have the time to learn and use the uClassify API, to simply copy and paste a HTML form snippet to any site.

We also want to publish some uClassify code libraries in different languages. One of our users, Alex Popescu, is currently implementing a Python library.

TrollGuard

In the beginning of January we released TrollGuard, a free WordPress spam filter plugin that uses uclassify.com. So far we have not had many users but recently it was featured on KillerStartups.com and since then we have an increasing number of users.

A very big thank you to Pacuraru Ovidiu who has given incredibly good feedback which led to an upgrade. We are looking forward to receive more feedback from all of you as our goal is to make the best spam filter plugin.

Using published classifiers

We’ve just implemented so that everyone with a uClassify account (free) can access public classifiers.

Once a classifier is published everyone can use it via the GUI or the web API and in return authors get a link to their website from everyone who use their classifiers. This should hopefully inspire more people to share their cool classifiers!

As an example of a published classifier check out the mood classifier by prfekt.se. Here is the list of all published classifiers.