gigi1111/alignment-models

Implementation of the Automatic Alignment Model and recipe data to align (data/alignment)

Automatic Alignment Model

Implementation of the automatic alignment model from the paper "Aligning Actions Across Recipe Graphs". In this paper we present two automatic alignment models (base, extended) and a simple baseline (cosine similarity).

Internal note: If you are searching for the version of the alignment model implied in our crowdsourcing task, please visit this private repository.

Requirements

You can find all the requirements in the file requirement.txt.

• Python 3.7
• Conllu 4.2.2
• Matplotlib 3.3.2
• Pandas 1.2.3
• Pytorch 1.7.1
• Transformers 4.0.0
• Flair 0.8.0.post1
• AllenNLP 0.9.0

Usage

Download the corpus from here into ./alignment-models/data/ folder for reproducing our experiment results. The data should be structured in one directory (/data), containing subdirectories corresponding to the different dishes, each of them including the recipes of the dish regrouped under a /recipes directory, and an "alignments.tsv" file. This file shows the crowsourced golden standard alignments of the correspoding recipes. Additionally, create the results folder where the trained models and their test results will be saved (Notes: You can change the hyperparameters and the path names in the file constants.py). Per default, the script looks for the following results folders:

Model Name	Saves To
Alignment Model (extended)	./results1
Alignment Model (base)	./results2
Cosine Similarity Baseline	./results3
Naive Model	./results4

To reproduce our experimental results, run the following command from this directory:

python main.py [model_name] --embedding_name [embedding_name]

where [model_name] could be one of the following:

• Sequence : Sequential Ordering of Alignments
• Cosine_similarity : Cosine model (Baseline)
• Naive : Common Action Pair Heuristics mode (Naive Model)
• Alignment-no-feature : Base Alignment model (w/o parent+child nodes)
• Alignment-with-feature : Extended Alignment model (with parent+child nodes)

and [embedding_name] could be one of the following:

• bert : BERT embeddings (default)
• elmo : ELMO embeddings

Results

Our experiment results are as follows:

Model Name	Accuracy
Sequential Order	16.5
Cosine Similarity	41.5
Naive Model	52.1
Alignment Model (base)	66.3
Alignment Model (extended)	72.4

Both the base and the extended Alignment models were trained for 10 folds each with 40 epochs.