Initial commit

.gitignore

@@ -0,0 +1,135 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# Elasticsearch data
+/0_webcrawler/data
+
+# scrapy test output file
+/0_webcrawler/test.json

.vscode/launch.json

@@ -0,0 +1,143 @@
+{
+    "version": "0.1.0",
+    "configurations": [
+        {
+            "name": "Scrapy to JSON",
+            "type": "python",
+            "request": "launch",
+            "module": "scrapy",
+            "args": [
+                "runspider",
+                "${file}",
+                "-a",
+                "start_url=https://kmi.fbi.h-da.de/aboutkmi.html",
+                "-o",
+                "test.json"
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Scrapy to ElasticSearch",
+            "type": "python",
+            "request": "launch",
+            "module": "scrapy",
+            "args": [
+                "runspider",
+                "${file}",
+                "-a",
+                "start_url=https://kmi.fbi.h-da.de/aboutkmi.html",
+                "-s",
+                "ELASTICSEARCH_INDEX=test123",
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Paraphrase file transformer",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "args": [
+                "--strategy",
+                "transformer_model",
+                "--file",
+                "training_examples.txt",
+                "--out",
+                "paraphrases1.xlsx"
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Paraphrase file backtranslation",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "args": [
+                "--strategy",
+                "backtranslation",
+                "--file",
+                "training_examples.txt",
+                "--out",
+                "paraphrases1.xlsx"
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Generate questions from website (answer-agnostic)",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "args": [
+                "--qg-strategy",
+                "answer_agnostic",
+                "--textblock-strategy",
+                "all",
+                "--index-name",
+                "kmi_crawler_2364dfd1-a060-44d9-85a8-eb7e5ecc3fe1",
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Generate questions from website (answer-aware)",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "args": [
+                "--qg-strategy",
+                "answer_aware",
+                "--textblock-strategy",
+                "all",
+                "--index-name",
+                "kmi_crawler_2364dfd1-a060-44d9-85a8-eb7e5ecc3fe1",
+                "--ner-library",
+                "flair",
+                "--ner-highlighting-strategy",
+                "word"
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Generate questions from website (answer-agnostic, translation pipeline)",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "args": [
+                "--qg-strategy",
+                "answer_agnostic",
+                "--textblock-strategy",
+                "all",
+                "--index-name",
+                "kmi_crawler_2364dfd1-a060-44d9-85a8-eb7e5ecc3fe1",
+                "--translation-pipeline",
+                "True"
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+        {
+            "name": "Generate questions from website (answer-agnostic faq random samples)",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "args": [
+                "--qg-strategy",
+                "answer_agnostic",
+                "--textblock-strategy",
+                "faq",
+                "--index-name",
+                "rki_crawler_57b57991-2755-4b68-9788-b382bf3c8196",
+                "--random-samples",
+                "True",
+                "--random-samples-count",
+                "3"
+            ],
+            "console": "integratedTerminal",
+            "cwd": "${fileDirname}",
+        },
+    ]
+}

0_webcrawler/README.md

@@ -0,0 +1,81 @@
+# Webcrawler
+The webcrawler extracts paragraphs from the target websites which can be used for question generation later on. Each distinct website is saved on an elasticsearch index for the given url. The name of the index and start URL are parameters for the crawler (see [Start the crawler from CLI](#Start-the-crawler-from-CLI)). 
+
+<img src="../images/crawling-websites-for-text.png" width="600" height="350" />
+
+## Webpage to text test
+
+|Task| |
+|---|---|
+|Webpage to text test|[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/TiloMichel/textgen-for-chatbot-training-german/blob/main/0_webcrawler/webpage_to_text_test.ipynb) |
+
+## Install
+Install all with
+
+`pip install -r requirements.txt`
+
+Scrapy, a fast high-level web crawling & scraping framework for Python https://github.com/scrapy/scrapy
+
+`pip install scrapy==2.5.1`
+
+Splash is a "Lightweight, scriptable browser as a service with an HTTP API" https://github.com/scrapinghub/splash
+
+`pip install scrapy_splash==0.8.0`
+
+Free and Open, Distributed, RESTful Search Engine https://github.com/elastic/elasticsearch
+
+`pip install elasticsearch==7.17.0`
+
+Heuristic based boilerplate removal tool https://github.com/miso-belica/jusText
+
+`pip install justext==3.0.0`
+
+### Install & run splash and Elasticsearch
+
+`docker-compose up`
+
+or
+
+`docker-compose -p "webcrawler" up`
+
+Elasticsearch server: http://localhost:9200/
+
+Splash frontend: http://localhost:8050/
+
+Recommended frontend for elasticsearch (browser extension): https://elasticvue.com/
+
+## Start the crawler from CLI
+The webcrawler can be started from CLI.
+
+Options
+* -s: Settings
+* -a: Arguments
+* -o: Output
+
+`scrapy crawl intent-finder-bot -s ELASTICSEARCH_INDEX=test -a start_url='https://kmi.fbi.h-da.de/aboutkmi.html' -o test.json`
+
+
+## Test URLs
+
+### No JavaScript content
+https://kmi.fbi.h-da.de/aboutkmi.html
+
+https://www.rki.de/DE/Home/homepage_node.html
+
+https://www.dge.de/wissenschaft/faqs/
+
+https://quotes.toscrape.com/
+
+https://books.toscrape.com/
+
+### JavaScript content
+http://www.webscrapingfordatascience.com/simplejavascript/
+
+https://quotes.toscrape.com/js/
+
+### Known problems
+- Cookie banner content is scraped. 
+- If content is rendered after mandatory user interaction it isn't scraped e. g. the cookie banner has to be accepted for content to render.
+
+## Further possible improvements
+- Run spiders on scrapyd (scrapy daemon service) for maintainability, monitoring and scalability https://scrapyd.readthedocs.io/en/stable/

0_webcrawler/docker-compose.yml

@@ -0,0 +1,29 @@
+version: '3'
+services:
+    # search engine
+    elasticsearch:
+        image: docker.elastic.co/elasticsearch/elasticsearch:7.17.0
+        container_name: elasticsearch
+        environment:
+            - discovery.type=single-node
+            - http.port=9200
+            - http.cors.enabled=true
+            - http.cors.allow-origin=/.*/
+            - http.cors.allow-headers=X-Requested-With,X-Auth-Token,Content-Type,Content-Length,Authorization
+            - http.cors.allow-credentials=true
+            - bootstrap.memory_lock=true
+            - 'ES_JAVA_OPTS=-Xms512m -Xmx512m'
+        ports:
+            - '9200:9200'
+            - '9300:9300'
+        ulimits:
+            memlock:
+                soft: -1
+                hard: -1
+        volumes:
+            - ./data:/usr/share/elasticsearch/data
+    splash:
+        image: scrapinghub/splash
+        container_name: splash
+        ports:
+            - 8050:8050