Hi Gunn, Here's a simple test case that would be useful to see the results of... Below I'm providing the output from the "short form" of the input name... I'll do the same with the long form too...
input : this is a test of mine i want to test many things with this this is a good day for testing i think i will test many things I have a test of mine that I went to test count.pl --ngram 3 test.cnt input statistic.pl --ngram 3 pmi test.pmi test.cnt statistic.pl --ngram 3 ll test.ll test.cnt statistic.pl --ngram 3 ps test.ps test.cnt statistic.pl --ngram 3 tmi test.tmi test.cnt marimba(46): more test.pmi 37 good<>day<>for<>1 10.4189 1 1 1 1 1 1 1 day<>for<>testing<>1 10.4189 1 1 1 1 1 1 1 that<>I<>went<>2 9.4189 1 1 2 1 1 1 1 a<>good<>day<>3 8.8339 1 3 1 1 1 1 1 think<>i<>will<>3 8.8339 1 1 3 1 1 1 1 for<>testing<>i<>3 8.8339 1 1 1 3 1 1 1 testing<>i<>think<>3 8.8339 1 1 3 1 1 1 1 with<>this<>this<>4 8.4189 1 1 2 2 1 1 1 many<>things<>with<>4 8.4189 1 2 2 1 2 1 1 of<>mine<>that<>4 8.4189 1 2 2 1 2 1 1 I<>went<>to<>4 8.4189 1 2 1 2 1 1 1 things<>I<>have<>4 8.4189 1 2 2 1 1 1 1 things<>with<>this<>4 8.4189 1 2 1 2 1 1 1 mine<>that<>I<>4 8.4189 1 2 1 2 1 1 1 i<>want<>to<>5 7.8339 1 3 1 2 1 1 1 is<>a<>good<>5 7.8339 1 2 3 1 2 1 1 I<>have<>a<>5 7.8339 1 2 1 3 1 1 1 mine<>i<>want<>5 7.8339 1 2 3 1 1 1 1 this<>this<>is<>5 7.8339 1 3 2 1 1 1 1 test<>many<>things<>6 7.4189 2 4 2 2 2 2 2 test<>of<>mine<>6 7.4189 2 4 2 2 2 2 2 to<>test<>many<>6 7.4189 1 1 4 2 1 1 2 many<>things<>I<>6 7.4189 1 2 2 2 2 1 1 will<>test<>many<>6 7.4189 1 1 4 2 1 1 2 this<>is<>a<>7 7.2490 2 3 2 3 2 2 2 i<>think<>i<>7 7.2490 1 3 1 3 1 1 1 want<>to<>test<>8 7.0970 1 1 2 5 1 1 2 went<>to<>test<>8 7.0970 1 1 2 5 1 1 2 a<>test<>of<>9 6.8339 2 3 4 2 2 2 2 of<>mine<>i<>9 6.8339 1 2 2 3 2 1 1 have<>a<>test<>10 6.5120 1 1 3 5 1 1 2 i<>will<>test<>10 6.5120 1 3 1 5 1 1 1 is<>a<>test<>11 5.5120 1 2 3 5 2 1 2 37 good<>day<>for<>1 10.4189 1 1 1 1 1 1 1 day<>for<>testing<>1 10.4189 1 1 1 1 1 1 1 that<>I<>went<>2 9.4189 1 1 2 1 1 1 1 a<>good<>day<>3 8.8339 1 3 1 1 1 1 1 think<>i<>will<>3 8.8339 1 1 3 1 1 1 1 for<>testing<>i<>3 8.8339 1 1 1 3 1 1 1 testing<>i<>think<>3 8.8339 1 1 3 1 1 1 1 with<>this<>this<>4 8.4189 1 1 2 2 1 1 1 many<>things<>with<>4 8.4189 1 2 2 1 2 1 1 of<>mine<>that<>4 8.4189 1 2 2 1 2 1 1 I<>went<>to<>4 8.4189 1 2 1 2 1 1 1 things<>I<>have<>4 8.4189 1 2 2 1 1 1 1 things<>with<>this<>4 8.4189 1 2 1 2 1 1 1 mine<>that<>I<>4 8.4189 1 2 1 2 1 1 1 i<>want<>to<>5 7.8339 1 3 1 2 1 1 1 is<>a<>good<>5 7.8339 1 2 3 1 2 1 1 I<>have<>a<>5 7.8339 1 2 1 3 1 1 1 mine<>i<>want<>5 7.8339 1 2 3 1 1 1 1 this<>this<>is<>5 7.8339 1 3 2 1 1 1 1 test<>many<>things<>6 7.4189 2 4 2 2 2 2 2 test<>of<>mine<>6 7.4189 2 4 2 2 2 2 2 to<>test<>many<>6 7.4189 1 1 4 2 1 1 2 many<>things<>I<>6 7.4189 1 2 2 2 2 1 1 will<>test<>many<>6 7.4189 1 1 4 2 1 1 2 this<>is<>a<>7 7.2490 2 3 2 3 2 2 2 i<>think<>i<>7 7.2490 1 3 1 3 1 1 1 want<>to<>test<>8 7.0970 1 1 2 5 1 1 2 went<>to<>test<>8 7.0970 1 1 2 5 1 1 2 a<>test<>of<>9 6.8339 2 3 4 2 2 2 2 of<>mine<>i<>9 6.8339 1 2 2 3 2 1 1 have<>a<>test<>10 6.5120 1 1 3 5 1 1 2 i<>will<>test<>10 6.5120 1 3 1 5 1 1 1 is<>a<>test<>11 5.5120 1 2 3 5 2 1 2 marimba(48): more test.ps 37 test<>many<>things<>1 8.2848 2 4 2 2 2 2 2 test<>of<>mine<>1 8.2848 2 4 2 2 2 2 2 this<>is<>a<>2 8.0492 2 3 2 3 2 2 2 a<>test<>of<>3 7.4739 2 3 4 2 2 2 2 good<>day<>for<>4 6.2218 1 1 1 1 1 1 1 day<>for<>testing<>4 6.2218 1 1 1 1 1 1 1 that<>I<>went<>5 5.5287 1 1 2 1 1 1 1 a<>good<>day<>6 5.1232 1 3 1 1 1 1 1 think<>i<>will<>6 5.1232 1 1 3 1 1 1 1 for<>testing<>i<>6 5.1232 1 1 1 3 1 1 1 testing<>i<>think<>6 5.1232 1 1 3 1 1 1 1 with<>this<>this<>7 4.8355 1 1 2 2 1 1 1 many<>things<>with<>7 4.8355 1 2 2 1 2 1 1 of<>mine<>that<>7 4.8355 1 2 2 1 2 1 1 I<>went<>to<>7 4.8355 1 2 1 2 1 1 1 things<>I<>have<>7 4.8355 1 2 2 1 1 1 1 things<>with<>this<>7 4.8355 1 2 1 2 1 1 1 mine<>that<>I<>7 4.8355 1 2 1 2 1 1 1 i<>want<>to<>8 4.4301 1 3 1 2 1 1 1 is<>a<>good<>8 4.4301 1 2 3 1 2 1 1 I<>have<>a<>8 4.4301 1 2 1 3 1 1 1 mine<>i<>want<>8 4.4301 1 2 3 1 1 1 1 this<>this<>is<>8 4.4301 1 3 2 1 1 1 1 to<>test<>many<>9 4.1424 1 1 4 2 1 1 2 many<>things<>I<>9 4.1424 1 2 2 2 2 1 1 will<>test<>many<>9 4.1424 1 1 4 2 1 1 2 i<>think<>i<>10 4.0246 1 3 1 3 1 1 1 want<>to<>test<>11 3.9193 1 1 2 5 1 1 2 went<>to<>test<>11 3.9193 1 1 2 5 1 1 2 of<>mine<>i<>12 3.7369 1 2 2 3 2 1 1 have<>a<>test<>13 3.5138 1 1 3 5 1 1 2 i<>will<>test<>13 3.5138 1 3 1 5 1 1 1 is<>a<>test<>14 2.8206 1 2 3 5 2 1 2 marimba(51): more test.tmi 37 test<>many<>things<>1 0.4986 2 4 2 2 2 2 2 test<>of<>mine<>1 0.4986 2 4 2 2 2 2 2 this<>is<>a<>2 0.4590 2 3 2 3 2 2 2 many<>things<>with<>3 0.4286 1 2 2 1 2 1 1 of<>mine<>that<>3 0.4286 1 2 2 1 2 1 1 a<>test<>of<>4 0.4265 2 3 4 2 2 2 2 many<>things<>I<>5 0.3756 1 2 2 2 2 1 1 good<>day<>for<>6 0.3585 1 1 1 1 1 1 1 day<>for<>testing<>6 0.3585 1 1 1 1 1 1 1 of<>mine<>i<>7 0.3564 1 2 2 3 2 1 1 is<>a<>good<>8 0.3541 1 2 3 1 2 1 1 is<>a<>test<>9 0.3506 1 2 3 5 2 1 2 to<>test<>many<>10 0.3205 1 1 4 2 1 1 2 will<>test<>many<>10 0.3205 1 1 4 2 1 1 2 that<>I<>went<>11 0.3045 1 1 2 1 1 1 1 want<>to<>test<>12 0.2974 1 1 2 5 1 1 2 went<>to<>test<>12 0.2974 1 1 2 5 1 1 2 a<>good<>day<>13 0.2841 1 3 1 1 1 1 1 think<>i<>will<>13 0.2841 1 1 3 1 1 1 1 for<>testing<>i<>13 0.2841 1 1 1 3 1 1 1 testing<>i<>think<>13 0.2841 1 1 3 1 1 1 1 with<>this<>this<>14 0.2515 1 1 2 2 1 1 1 I<>went<>to<>14 0.2515 1 2 1 2 1 1 1 things<>I<>have<>14 0.2515 1 2 2 1 1 1 1 things<>with<>this<>14 0.2515 1 2 1 2 1 1 1 mine<>that<>I<>14 0.2515 1 2 1 2 1 1 1 i<>want<>to<>15 0.2323 1 3 1 2 1 1 1 I<>have<>a<>15 0.2323 1 2 1 3 1 1 1 mine<>i<>want<>15 0.2323 1 2 3 1 1 1 1 this<>this<>is<>15 0.2323 1 3 2 1 1 1 1 have<>a<>test<>16 0.2265 1 1 3 5 1 1 2 i<>think<>i<>17 0.2142 1 3 1 3 1 1 1 i<>will<>test<>18 0.1960 1 3 1 5 1 1 1 On Thu, Jun 11, 2009 at 6:27 AM, gunnlyse<gunnl...@yahoo.no> wrote: > > > Hello, > thank you for this clarification! > > The NSP values still do not match mine, and I see that it concerns ll, pmi, > ps as well as tmi for trigrams. Evidently, there must be some error which > probably lies in the observed or estimated frequencies (since all four > measures produce different results than mine) > > I need to ask for two clarifications: > > (1) estimated frequency: The webpage/pmi file says: > n1pp * np1p * npp1 > m111= -------------------- > nppp > but the file 3D.pm says > $m111=$n1pp*$np1p*$npp1/($nppp**2); " > > which I take to mean that we use, not nppp, but the exponent: > n1pp * np1p * npp1 > m111= -------------------- > nppp * nppp > If so, which one sould I really use? > > (2) Furthermore, let us return to the example trigram. When I compute the > example trigram's pmi in the way I understand the code, I get the value > -15.24452, instead of the NSP package's 6.4127. > All the observed frequencies needed for pmi are directly available in the > example trigram line, so the only thing that can explain diverging results > is HOW we compute the value. > May I therefore ask if you agree with the way I understand the code? > > For the trigram > 355663266 > at<>det<>er<>262744 7073841 9391062 5872364 1234064 647295 1064083 > > I compute m111 as: >> > m111= 7073841 * 9391062 * 5872364 >> > ------------------------- >> > 355663266 >> > >> > = 1.0968417e+12 >> > >> > and PMI = log (262744 / 1.0968417e+12) = -15.24452 > >> > NSP's pmi returns (using the command line: >> > statistic.pl --ngram 3 pmi outputfile inputfile ) >> > produces the following line > at<>det<>er<>1 6.4127 262744 7073841 9391062 5872364 1234064 647295 1064083 > > Best, > Gunn > > -- Ted Pedersen http://www.d.umn.edu/~tpederse
