Dimas263 / NLP_NER_BILSTM_Named_Entity_Recognition

NLP Named Entity Recognition dalam bidang Biomedis, mendeteksi teks dan membuat klasifikasi apakah teks tersebut mempunyai entitas plant atau disease, memberi label pada teks, menguji hubungan entitas plant dan disease, menilai kecocokan antara kedua entitas, membandingkan hasil uji dengan menggunakan models BILSTM

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

NLP

Named Entity Recognition (NER) - BILSTM

Slamet Riyanto S.Kom., M.M.S.I.

Dimas Dwi Putra

Architecture

Sentence # Word POS Tag
Sentence: 0 studies NNS O
Sentence: 0 on IN O
Sentence: 0 magnesium NN O
Sentence: 0 s NN O
Sentence: 0 mechanism NN O
Sentence: 0 of IN O
Sentence: 0 action NN O
Sentence: 0 in IN O
Sentence: 0 digitalis NN B-plant
Sentence: 0 induced VBD O
Sentence: 0 arrhythmias NNS B-disease
...
Entities precision recall f1-score support excecution time processor ram model batch size epochs
PAD 0,976923 0,976923 0,976923 130 0.06.21 CPU High 1 16 20
Disease 0,460432 0,450704 0,455516 142
Plant 0,854962 0,829630 0,842105 135
micro avg 0,757500 0,744472 0,750929 407
macro avg 0,764106 0,752419 0,758181 407
weighted avg 0,756268 0,744472 0,750289 407
F-1 Scores 75,10%

Predict

Sample number 2 of 131 (Test Set)
Word           ||True ||Pred
==============================
background     : O     O
and            : O     O
aims           : O     O
large          : O     O
scale          : O     O
epidemiological: O     O
studies        : O     O
have           : O     O
shown          : O     O
that           : O     O
drinking       : O     O
more           : O     O
than           : O     O
two            : O     O
cups           : O     O
of             : O     O
coffee         : plant plant
per            : O     O
day            : O     O
reduces        : O     O
the            : O     O
risks          : O     O
hepatitis      : disease disease
liver          : O     disease

Save output model as .hdf5

Other Content

Websites Prediction

Named Entity Recognition (NER)

Relation Extraction (RE)

About

NLP Named Entity Recognition dalam bidang Biomedis, mendeteksi teks dan membuat klasifikasi apakah teks tersebut mempunyai entitas plant atau disease, memberi label pada teks, menguji hubungan entitas plant dan disease, menilai kecocokan antara kedua entitas, membandingkan hasil uji dengan menggunakan models BILSTM


Languages

Language:Jupyter Notebook 100.0%