Dependency parsing with multiple texts

In this exercise, you will apply the dependency parsing and CoNLL-U procedures from the previous modules to multiple texts. Use the same five sample texts from the English POS tagging exercises. Follow the same steps from that activity to load the .txt files and process them with nlp_en.pipe() so that you have:

text_files
docs

You do not need to download the sample corpus again, if you already saved it in your local computer.

Inspect dependency annotations across texts

Choose one of the sample texts and inspect its dependency annotations:

doc = docs[0]

for token in doc:
    if not token.is_space:
        print(
            token.text,
            token.pos_,
            token.head.text,
            token.dep_,
            sep="\t"
        )

Try this with different sample texts. Pay particular attention to the syntactic head and dependency relation assigned to each token.

Create CoNLL-U files

Use the doc_to_conllu() function from the CoNLL-U format module to convert each sample text into CoNLL-U format.

First, create a folder for the output files:

from pathlib import Path

output_folder = Path("dependency_output")
output_folder.mkdir(parents=True, exist_ok=True)

Then convert and save all five texts:

for text_file, doc in zip(text_files, docs):

    conllu_text = doc_to_conllu(
        doc,
        text_label=text_file.stem
    )

    output_path = (
        output_folder /
        f"{text_file.stem}.conllu"
    )

    output_path.write_text(
        conllu_text,
        encoding="utf-8"
    )

    print(f"Saved: {output_path.name}")

Your output folder should contain:

sample1.conllu
sample2.conllu
sample3.conllu
sample4.conllu
sample5.conllu

Check your output

A .conllu file is a plain-text file, so you can open it with any text editor. For example:

VSCode: right-click the file and select Open With → Text Editor
Excel: open or import the .conllu file as a tab-delimited text file. This is useful for viewing the ten CoNLL-U fields as separate columns.
macOS: open it with TextEdit
Windows: open it with Notepad?

You can also inspect the file directly in Python:

conllu_file = output_folder / "sample1.conllu"

print(
    conllu_file.read_text(
        encoding="utf-8"
    )
)

Open at least one .conllu file and check that:

each sentence includes # sent_id and # text;
each token is represented using the ten CoNLL-U columns;
HEAD contains the syntactic head ID;
DEPREL contains the dependency relation;
sentences are separated by a blank line.

Submission

Submit the five .conllu files in a single compressed folder.

Use the following filename:

lastname_firstname_dependency_analysis.zip

Back to Module 8