matsimilarity.core package

Submodules

matsimilarity.core.SimilarityMeasure module

MAT-Tools: Python Framework for Multiple Aspect Trajectory Data Mining

The present package offers a tool, to support the user in the task of data analysis of multiple aspect trajectories. It integrates into a unique framework for multiple aspects trajectories and in general for multidimensional sequence data mining methods. Copyright (C) 2022, MIT license (this portion of code is subject to licensing from source project distribution)

Authors: - Vanessa Lago Machado - Tarlis Portela

class matsimilarity.core.SimilarityMeasure.SimilarityMeasure(dataset_descriptor: DataDescriptor | None = None)[source]

Bases: ABC

property attributes

Getter for attributes from the data descriptor.

Returns:: List of attributes from the data descriptor.
Return type:: List[FeatureDescriptor]

similarity(t1: MultipleAspectSequence, t2: MultipleAspectSequence) → float[source]

Computes the similarity score of the given MAT.

Parameters:

t1 (MultipleAspectSequence instance of the trajectory 1.)
t2 (MultipleAspectSequence instance of the trajectory 2.)

Returns:

score – Similarity score (between 0 and 1).

Return type:

float

matsimilarity.core.utils module

MAT-Tools: Python Framework for Multiple Aspect Trajectory Data Mining

The present package offers a tool, to support the user in the task of data analysis of multiple aspect trajectories. It integrates into a unique framework for multiple aspects trajectories and in general for multidimensional sequence data mining methods. Copyright (C) 2022, MIT license (this portion of code is subject to licensing from source project distribution)

Authors: - Vanessa Lago Machado - Tarlis Portela

matsimilarity.core.utils.similarity_matrix(A, B=None, measure=None, n_jobs=1)[source]

Computes the similarity matrix from a list of trajectories Ta x Ta, or Ta x Tb (if provided).

Parameters:

Alist of MultipleAspectSequence: List of Trajectory objects to compute similarity from. Each trajectory should be a MultipleAspectSequence.
Blist of MultipleAspectSequence (optional): List of Trajectory objects to compute similarity to A. Each trajectory should be a MultipleAspectSequence.
measureSimilarityMeasure instance: A class with a similarity function that takes two trajectories and returns a similarity score.
n_jobsint, optional: The number of parallel jobs to use for computation (default is 1).

Returns:

np.ndarraysimilarity array with shape (len(A), len(B)).: A 2D numpy array containing similarity scores between trajectories. The element at [i, j] represents the similarity between trajectory A[i] and B[j].

Example:

>>> T = [Trajectory1, Trajectory2, Trajectory3]
>>> sim_matrix = similarity_matrix(T, measure=MUITAS(), n_jobs=4)
>>> print(sim_matrix)
[[1.0, 0.8, 0.3],
 [0.8, 1.0, 0.5],
 [0.3, 0.5, 1.0]]

Source:

From trajminer with MIT License: https://github.com/trajminer/trajminer/blob/master/trajminer/similarity/pairwise.py