News for hackers. Stuff that matters.
Re:Sequence Tagging with Tensorflow by gajju3588 on Wednesday March 14, 2018 @05:59AM (#16583525)
Nice article. Explains the whole things really well. I have found LSTM is little overkill for NER, as your entities are limited and data will not be very huge. CRF and Max-Ent seems to better options.

People who want to try it out on some real dataset, can get a sample data for NER from here :

https://dataturks.com/projects/Mohan/Best%20Buy%20E-commerce...

Re:Sequence Tagging with Tensorflow by Lindrian on Wednesday March 14, 2018 @06:42AM (#16583705)
Don't agree. LSTM is a very good option for NER. I have personally experimented with simple FFNN and language models to accommodate variable width input and achieved very good results. I will be publishing a paper on this soon.

The system in the article fails at quite simple sentences, such as:

"Rex Tillerson is the secretary of state."

It tags "Rex Tillerson" as ORG instead of PER. If you change "is" to "was" it tags it as PER.

Re:Sequence Tagging with Tensorflow by melzarei on Wednesday March 14, 2018 @07:00AM (#16583777)
Disclaimer: as you may notice, the tagger is far from being perfect. Some errors are due to the fact that the demo uses a reduced vocabulary (lighter for the API). But not all. It is also very sensible to capital letters, which comes both from the architecture of the model and the training data. For more information about the demo, see here.
Re:Sequence Tagging with Tensorflow by bravura on Wednesday March 14, 2018 @12:38PM (#16585996)
"I have found LSTM is little overkill for NER, as your entities are limited and data will not be very huge."

Incorrect. Inducing word representations on a huge corpus, like Common Crawl, can be more effective than manually building a small Gazeeter of named entities.

Re:Sequence Tagging with Tensorflow by pacala on Wednesday March 14, 2018 @04:15PM (#16587920)
One can do semi-supervised training with LSTMs, using http://allennlp.org/elmo or similar. That may use huge amounts of data, for which MaxEnt or CRF are not sufficient.