Linearisation in Matxin - Revision history

Francis Tyers at 19:03, 29 November 2016

2016-11-29T19:03:26Z

Bujol at 18:40, 29 November 2016

2016-11-29T18:40:50Z

Francis Tyers at 14:24, 29 November 2016

2016-11-29T14:24:32Z

Francis Tyers at 22:58, 16 November 2016

2016-11-16T22:58:52Z

Francis Tyers: Created page with " ==Thoughts==

                    ..."

2016-11-14T20:20:21Z

Created page with " ==Thoughts== <pre> <NODE ord="3" mi="n|sg"> <NODE ord="2" si="amod"/> <NODE ord="1" si="det"/> </NODE> <NODE ord="2" si="root""> <NODE ord="1" si="nsubj"/> ..."

New page

==Thoughts==

<pre>
<NODE ord="3" mi="n|sg">
<NODE ord="2" si="amod"/>
<NODE ord="1" si="det"/>
</NODE>

<NODE ord="2" si="root"">
<NODE ord="1" si="nsubj"/>
<NODE ord="3" si="dobj"/>
</NODE>
</pre>

==Relevant papers==

* http://www.anthology.aclweb.org/C/C04/C04-1090.pdf

[[Category:Matxin]]

@@ Line 1: / Line 1: @@
+{{TOCD}}
+==Paper reviews==
-==Linearisation of Dependency Trees - review==
 * Transition-Based Syntactic Linearisation with Lookahead Features” - Ratish Puduppully, Yue Zhang, Manish Shrivastava - [http://www.aclweb.org/anthology/N/N16/N16-1058.pdf]

@@ Line 1: / Line 1: @@
+==Linearisation of Dependency Trees - review==
+* Transition-Based Syntactic Linearisation with Lookahead Features” - Ratish Puduppully, Yue Zhang, Manish Shrivastava - [http://www.aclweb.org/anthology/N/N16/N16-1058.pdf]
+<pre>
+The brand new algorithm. Has the best results that have been seen. 30x faster than the next one. Has published the code on Github.
+“Our system improves upon the previous best scores by 8.7 BLEU points for the task of unlabeled syntactic linearisation.
+For the task of labeled syntactic linearisation, we achieve the score of 91.8 BLEU points, the highest results reported so far.”
+Unfortunately, tested only for English.
+</pre>
+* “Partial-Tree Linearisation: Generalized Word Ordering for Text Synthesis” - Yue Zhang - [http://people.sutd.edu.sg/~yue_zhang/pub/ijcai13.pdf]
+<pre>
+The best algorithm of its time, yet it was outperformed by the next algorithms (developed by the same author).
+It has good results in sensible time constraints (5s), but it uses a lot of memory for the training set
+(tens of thousands of examples needed) and the training time is also long.
+Accuracy: “Our system gave a BLEU score of 89.3, comparable to the best-performing system of the shared task,
+although our results are not directly comparable because we used the PTB POS set and the Penn2Malt dependency labels,
+which are less fine-grained than the shared task data.” That’s the best result there was at that time, everything was performed in the 5s time constraints.
+Tried only in English
+</pre>
+* “Transition-Based Syntactic Linearisation” - Yijia Liu , Yue Zhang , Wanxiang Che , Bing Qin [http://people.sutd.edu.sg/~yue_zhang/pub/naacl15.yijia.pdf]
+<pre>
+Faster and more precise than the last one. Moreover, they even share their code for the algorithm, which would surely help during the implementation.
+Again, the only problem is the need for a big learning set.
+Got 81 on the BLEU score, yet compared to the one above it performed better.
+Only English (Wall Street Journal)
+</pre>
+* “Experiments with Generative Models for Dependency Tree Linearisation” - Richard Futrell and Edward Gibson - [https://www.aclweb.org/anthology/D15-1231]
+<pre>
+Good algorithm, doesn’t require that much of a memory, but no mention of the performance time-wise (probably it was poor).
+Also, when it comes to the accuracy, it was outperformed by the previous one.
+Got 58 BLEU (some random techniques in other papers got 38, so quite bad).
+But did it for many languages: Basque, Czech, English, Finnish, French, German, Hebrew, Indonesian, Persian, Swedish, Spanish
+</pre>
+* “Sentence Realisation with Unlexicalized Tree Linearisation Grammars” Rui WANG Yi ZHANG  - [https://www.aclweb.org/anthology/C12-2127]
+<pre>
+Again, another algorithm by the same author. It again requires a lot of memory for training.
+But it’s checked on many languages (BLEU score) : Catalan (76), Chinese (81), Czech (67), English (85), German (74), Spanish (73)
+</pre>
+* “Generating Non-Projective Word Order in Statistical Linearisation” - Bernd Bohnet Anders Bjorkelund Jonas Kuhn Wolfgang Seeker Sina Zarrieß - [https://aclweb.org/anthology/D/D12/D12-1085.pdf]
+<pre>
+Nice results, everything neatly explained, surprisingly good score!
+checked on  (BLEU score) Dutch (82) , Danish (86), Hungarian(77), Czech (72), English(93) and German(80).
+</pre>
 ==Thoughts==

@@ Line 18: / Line 18: @@
 * http://www.anthology.aclweb.org/C/C04/C04-1090.pdf
+* http://www.coli.uni-saarland.de/~rwang/pubs/COLING2012.pdf
+* http://olst.ling.umontreal.ca/pdf/kahane+lareau-esslli16.pdf

@@ Line 14: / Line 14: @@
 </NODE>
 </pre>
+==External resources==
+* [https://github.com/SUTDNLP/ZGen ZGen:  ZGen is a linearisation system that constructs natural language sentences from bags of words]
 ==Relevant papers==