Why model is replacing Proper noun to pronoun? E.g "Oxygen" to "It"

[bhavikapanara]

The model is replacing proper noun to the pronoun. For example,

Here is the input statement: oxygen is a chemical element with symbol o and atomic number 8 .

The output is: It has the chemical symbol o . It has the atomic number 8 .

The expected output should be like this: oxygen is a chemical element with symbol o or oxygen has the chemical symbol o oxygen has the atomic number 8

How to stop such replacement? Can anyone please help me to figure out this problem.

Thanks, Bhavika

[louismartin]

Hi,

Yes this is a case of failure of the model. It will probably hard to fix and would require a model for anaphora resolution which is out of the scope of this repo sadly.

Good luck, Louis

[bhavikapanara]

Thanks, @louismartin for the quick response.

After, reading your paper, I came to know that there are four parameters you need to adjust. I have also played with this but, nothing came out of it.

# Load best model
best_model_dir = prepare_models()
recommended_preprocessors_kwargs = {
    'LengthRatioPreprocessor': {'target_ratio': 0.90},
    'LevenshteinPreprocessor': {'target_ratio': 0.75},
    'WordRankRatioPreprocessor': {'target_ratio': 0.75},
    'DependencyTreeDepthRatioPreprocessor':{'target_ratio':0.75},
    'SentencePiecePreprocessor': {'vocab_size': 10000},
}

Any suggestion on this.

[louismartin]

Hi, The control parameters will sadly not fix this problem, which is more due to what the neural network has learned during training and that can't really be changed.