Word | Frequency | Number of left neighbors | Ratio |
---|---|---|---|
Bu | 2454 | 6 | 409.0000 |
erkek | 329 | 1 | 329.0000 |
sorunlarınızı | 269 | 1 | 269.0000 |
düzleştirmek | 268 | 1 | 268.0000 |
Arıyorum | 468 | 2 | 234.0000 |
Daha | 224 | 1 | 224.0000 |
onlara | 207 | 1 | 207.0000 |
görünür | 207 | 1 | 207.0000 |
fotoğraflar | 203 | 1 | 203.0000 |
kullanıcılara | 202 | 1 | 202.0000 |
işaretlenmiş | 193 | 1 | 193.0000 |
görüntülemeyi | 192 | 1 | 192.0000 |
erotik | 192 | 1 | 192.0000 |
IBM | 175 | 1 | 175.0000 |
mesaj | 173 | 1 | 173.0000 |
kadın | 343 | 2 | 171.5000 |
artık | 164 | 1 | 164.0000 |
hem | 325 | 2 | 162.5000 |
size | 161 | 1 | 161.0000 |
oranını | 319 | 2 | 159.5000 |
We look at words having extremely few left neighbors. Because we have an artificial additional word before any sentence beginning, this is the only left neighbor for words appearing only (or mainly) at the beginning of sentences. In addition to the artificial sentence beginning word, there is a corresponding artificial sentence ending word. This is denoted by %$% and has as left neighbors mainly punctuation marks.
Table data:
select word, w.freq, count(c.w1_id), w.freq/count(c.w1_id) as r from words w, co_n c where w.w_id=w2_id and w1_id>100 group by w2_id order by r desc limit 20;
Diagram data:
select w.freq, count(c.w1_id) from words w, co_n c where w.w_id=w2_id and w1_id>100 group by w2_id;
In the above diagram seems to be a larger variation than in the diagram in the previous subsection. What is the reason?
There is no symmetry below and above average. Why?
5.1.7.1 Number of NN co-occurrences vs. Frequency I
5.1.7.3 Number of left vs. right NN co-occurrences