Pattern Locator KG

Finding statistical relational patterns

Workflow

## Description The Pattern Locator KG is a repository that contains a pattern matching algorithm that locates specific patterns in a given dataset from knowledge graphs triplets. Currently the project supports the following pattern extraction from KGs: - Reflection - Symmetry - Transitivity - Implication - Inverse - Composition

Installation

To install and run the program, follow these steps:

1. Clone this repository to your local machine.
2. Install the required dependencies by running the command pip install -r requirements.txt.
3. Run the program using the following command

python relational_pattern_lookout.py --data_dir <DATA_DIRECTORY> --data_name <DATASET_NAME> --file_name <FILE_NAME> --save_pattern_dir <DIRECTORY_NAME_FOR_SAVING_PATTERNS> 

Usage

The program takes a dataset and a knowledge graph as inputs, and outputs the locations of specific patterns within the dataset. To use the program, follow these steps:

1. Prepare your dataset and knowledge graph files in the appropriate format normally in a text set where we have {h,r,t}.
2. Run the program using the following command

python relational_pattern_lookout.py --data_dir <DATA_DIRECTORY> --data_name <DATASET_NAME> --file_name <FILE_NAME> --save_pattern_dir <DIRECTORY_NAME_FOR_SAVING_PATTERNS> 

Arguments:

DATA_DIRECTORY is the directory containing the dataset

DATASET_NAME is the name of the dataset file (including the extension)

FILE_NAME is the name of the file to save the identified patterns in (including the extension)

DIRECTORY_NAME_FOR_SAVING_PATTERNS is the directory name for saving the identified patterns. The program will output the detected patterns in the given location

General Structure of the Repository

.
├── data/
│   └── dataset1/
│   │   ├── saved_patterns/
│   │   │   ├── implication_patterns.txt
│   │   │   ├── inverse_patterns.txt
│   │   │   ├── reflection_patterns.txt
│   │   │   ├── symmetric_patterns.txt
│   │   │   ├── transitive_patterns.txt
│   │   │   └── composition_patterns.txt
│   │   └── train.txt 
│   └── dataset2/
│   │   └── ...
├── utils/
│   ├── utility_functions.py
│   └── ...
├── temporal_pattern/
│   ├── evaluation.py
│   ├── run.py
│   ├── temporal_pattern_lookout.py
│   ├── tpl.py
│   ├── utilities.py
│   └── ...
├── README.md
├── requirements.txt
├── .gitignore
├── relational_pattern_lookout.py
└── ...

Files for statistical pattern lookout

• relational_pattern_lookout.py: This is the main file for the statistical pattern lookout. It contains the main function that calls the other functions in the program.
• utility_functions.py: This file contains the utility functions (including the codes to find statistical patterns) used in the program.

Temporal Pattern Lookout:

• This script is to find pattern in (Temporal) Knowledge Graph. The file format is in quaternion (subject, predicate, object, time) or triples ((subject, predicate, object).
• Unzip the data.zip in the same folder as folder 'temporal_pattern_lookout' How to run: two hyperparameters needed to be determined, i.e. dataset and threshold for pattern lookup table. example:
  • python run.py --dataset icews15 --threshold 0.5

Contributing

If you want to contribute to this repository, please fork the repository, create a new branch, make your changes, and submit a pull request with a detailed description of your changes.

License

This repository is licensed under the MIT License.