import streamlit as st
import pandas as pd
# Custom CSS for better styling
st.markdown("""
""", unsafe_allow_html=True)
# Main Title
st.markdown('Detect 10 Different Entities in Hebrew (hebrew_cc_300d embeddings)
', unsafe_allow_html=True)
# Introduction
st.markdown("""
Named Entity Recognition (NER) models identify and categorize important entities in a text. This page details a word embeddings-based NER model for Hebrew texts, using the hebrew_cc_300d word embeddings. The model is pretrained and available for use with Spark NLP.
""", unsafe_allow_html=True)
# Model Description
st.markdown('Description
', unsafe_allow_html=True)
st.markdown("""
This model uses Hebrew word embeddings to find 10 different types of entities in Hebrew text. It is trained using hebrew_cc_300d word embeddings, so please use the same embeddings in the pipeline. It can identify the following types of entities:
- PERS (Persons)
- DATE (Dates)
- ORG (Organizations)
- LOC (Locations)
- PERCENT (Percentage)
- MONEY (Money)
- TIME (Time)
- MISC_AFF (Miscellaneous Affiliation)
- MISC_EVENT (Miscellaneous Event)
- MISC_ENT (Miscellaneous Entity)
""", unsafe_allow_html=True)
# Setup Instructions
st.markdown('Setup
', unsafe_allow_html=True)
st.markdown('To use the model, you need Spark NLP installed. You can install it using pip:
', unsafe_allow_html=True)
st.code("""
pip install spark-nlp
pip install pyspark
""", language="bash")
st.markdown("Then, import Spark NLP and start a Spark session:
", unsafe_allow_html=True)
st.code("""
import sparknlp
# Start Spark Session
spark = sparknlp.start()
""", language='python')
# Example Usage
st.markdown('Example Usage with Hebrew NER Model
', unsafe_allow_html=True)
st.markdown("""
Below is an example of how to set up and use the hebrewner_cc_300d model for named entity recognition in Hebrew:
""", unsafe_allow_html=True)
st.code('''
from sparknlp.base import *
from sparknlp.annotator import *
from pyspark.ml import Pipeline
from pyspark.sql.functions import col, expr, round, concat, lit, explode
# Define the components of the pipeline
documentAssembler = DocumentAssembler() \\
.setInputCol("text") \\
.setOutputCol("document")
sentence_detector = SentenceDetector() \\
.setInputCols(["document"]) \\
.setOutputCol("sentence")
tokenizer = Tokenizer() \\
.setInputCols(["sentence"]) \\
.setOutputCol("token")
word_embeddings = WordEmbeddingsModel.pretrained("hebrew_cc_300d", "he") \\
.setInputCols(["sentence", "token"]) \\
.setOutputCol("embeddings")
ner = NerDLModel.pretrained("hebrewner_cc_300d", "he") \\
.setInputCols(["sentence", "token", "embeddings"]) \\
.setOutputCol("ner")
ner_converter = NerConverter().setInputCols(["sentence", "token", "ner"]).setOutputCol("ner_chunk")
# Create the pipeline
pipeline = Pipeline(stages=[documentAssembler, sentence_detector, tokenizer, word_embeddings, ner, ner_converter])
# Create sample data
example = """
ב- 25 לאוגוסט עצר השב"כ את מוחמד אבו-ג'וייד , אזרח ירדני , שגויס לארגון הפת"ח והופעל על ידי חיזבאללה. אבו-ג'וייד התכוון להקים חוליות טרור בגדה ובקרב ערביי ישראל , לבצע פיגוע ברכבת ישראל בנהריה , לפגוע במטרות ישראליות בירדן ולחטוף חיילים כדי לשחרר אסירים ביטחוניים.
"""
data = spark.createDataFrame([[example]]).toDF("text")
# Fit and transform data with the pipeline
result = pipeline.fit(data).transform(data)
# Select the result, entity
result.select(
expr("explode(ner_chunk) as ner_chunk")
).select(
col("ner_chunk.result").alias("chunk"),
col("ner_chunk.metadata").getItem("entity").alias("ner_label")
).show(truncate=False)
''', language="python")
import pandas as pd
# Create the data for the DataFrame
data = {
"chunk": [
"25 לאוגוסט",
"השב\"כ",
"מוחמד אבו-ג'וייד",
"ירדני",
"הפת\"ח",
"חיזבאללה",
"אבו-ג'וייד",
"בגדה",
"ערביי",
"ישראל",
"ברכבת ישראל",
"בנהריה",
"ישראליות",
"בירדן"
],
"ner_label": [
"DATE",
"ORG",
"PERS",
"MISC_AFF",
"ORG",
"ORG",
"PERS",
"LOC",
"MISC_AFF",
"LOC",
"ORG",
"LOC",
"MISC_AFF",
"LOC"
]
}
# Creating the DataFrame
df = pd.DataFrame(data)
df.index += 1
st.dataframe(df)
# Model Information
st.markdown('Model Information
', unsafe_allow_html=True)
st.markdown("""
The hebrewner_cc_300d model details are as follows:
- Model Name: hebrewner_cc_300d
- Type: ner
- Compatibility: Spark NLP 4.0.2+
- License: Open Source
- Edition: Official
- Input Labels: [document, token, word_embeddings]
- Output Labels: [ner]
- Language: he
- Size: 14.8 MB
""", unsafe_allow_html=True)
# Benchmark Section
st.markdown('Benchmark
', unsafe_allow_html=True)
st.markdown("""
Evaluating the performance of NER models is crucial to understanding their effectiveness in real-world applications. Below are the benchmark results for the hebrewner_cc_300d model, focusing on various named entity categories. The metrics used include precision, recall, and F1-score, which are standard for evaluating classification models.
""", unsafe_allow_html=True)
st.markdown("""
---
| Label | TP | FP | FN | Precision | Recall | F1-Score |
|--------------|-----|-----|-----|-----------|---------|----------|
| I-TIME | 5 | 2 | 0 | 0.714286 | 1.000000| 0.833333 |
| I-MISC_AFF | 2 | 0 | 3 | 1.000000 | 0.400000| 0.571429 |
| B-MISC_EVENT | 7 | 0 | 1 | 1.000000 | 0.875000| 0.933333 |
| B-LOC | 180 | 24 | 37 | 0.882353 | 0.829493| 0.855107 |
| I-ORG | 124 | 47 | 38 | 0.725146 | 0.765432| 0.744745 |
| B-DATE | 50 | 4 | 7 | 0.925926 | 0.877193| 0.900901 |
| I-PERS | 157 | 10 | 15 | 0.940120 | 0.912791| 0.926254 |
| I-DATE | 39 | 7 | 8 | 0.847826 | 0.829787| 0.838710 |
| B-MISC_AFF | 132 | 11 | 9 | 0.923077 | 0.936170| 0.929577 |
| I-MISC_EVENT | 6 | 0 | 2 | 1.000000 | 0.750000| 0.857143 |
| B-TIME | 4 | 0 | 1 | 1.000000 | 0.800000| 0.888889 |
| I-PERCENT | 8 | 0 | 0 | 1.000000 | 1.000000| 1.000000 |
| I-MISC_ENT | 11 | 3 | 10 | 0.785714 | 0.523810| 0.628571 |
| B-MISC_ENT | 8 | 1 | 5 | 0.888889 | 0.615385| 0.727273 |
| I-LOC | 79 | 18 | 23 | 0.814433 | 0.774510| 0.793970 |
| B-PERS | 231 | 22 | 26 | 0.913044 | 0.898833| 0.905882 |
| B-MONEY | 36 | 2 | 2 | 0.947368 | 0.947368| 0.947368 |
| B-PERCENT | 28 | 3 | 0 | 0.903226 | 1.000000| 0.949152 |
| B-ORG | 166 | 41 | 37 | 0.801932 | 0.817734| 0.809756 |
| I-MONEY | 61 | 1 | 1 | 0.983871 | 0.983871| 0.983871 |
| Macro-average| 1334| 196 | 225 | 0.899861 | 0.826869| 0.861822 |
| Micro-average| 1334| 196 | 225 | 0.871895 | 0.855677| 0.863710 |
""", unsafe_allow_html=True)
# Summary
st.markdown('Summary
', unsafe_allow_html=True)
st.markdown("""
This page provided an overview of the hebrewner_cc_300d model for Hebrew NER. We discussed how to set up and use the model with Spark NLP, including example code and results. We also provided details on the model's specifications and links to relevant resources for further exploration.
""", unsafe_allow_html=True)
# References
st.markdown('References
', unsafe_allow_html=True)
st.markdown("""
""", unsafe_allow_html=True)
# Community & Support
st.markdown('Community & Support
', unsafe_allow_html=True)
st.markdown("""
""", unsafe_allow_html=True)