Categorical Similarity Space

Classes

CategoricalSimilaritySpace(category_input: superlinked.framework.common.schema.schema_object.StringList | superlinked.framework.common.schema.schema_object.String | list[superlinked.framework.common.schema.schema_object.String | superlinked.framework.common.schema.schema_object.StringList], categories: list[str], negative_filter: float = 0.0, uncategorized_as_category: bool = True) : Represents a space for encoding categorical similarity.

A CategoricalSimilaritySpace is designed to measure the similarity between items that
are grouped into a finite number of textual categories. The categories are represented
in an n-hot encoded vector, with the option to apply a negative filter for unmatched
categories, enhancing the distinction between matching and non-matching category items.
Negative_filter allows for filtering out unmatched categories, by setting
it to a large negative value, effectively resulting in large negative similarity between
non-matching category items. A category input not present in categories will be encoded
as `other` category. These categories will be similar to each other by default. Set
uncategorized_as_category parameter to False in order to suppress this behavior - this
way other categories are not similar to each other in any case - not even to the same
`other` category. To make that specific category value similar to only the same category
items, consider adding it to `categories`.

Attributes:
    category_input (StringList | String | list[String | StringList]):
        The schema field containing input categories to be considered in the similarity space.
        Input contains one or more categories in a list if `StringList` is provided.
        If `String` is provided, then the input must be a single value.
    categories (List[str]): A list of categories that defines the dimensionality of the
        one-hot encoded vector. Any category not listed is considered as 'other'.
    negative_filter (float): A value to represent unmatched categories in the one-hot vector.
        Instead of using 0, which typically represents the absence of a category, this allows
        for a different representation - resulting in effectively filtering out items that has
        non-matching categories.
    uncategorized_as_category (bool): If set to False, the similarity between other categories will be
        set to 0, or negative_filter if set. By this we can control if a category_input not in
        categories will be similar to other category_inputs not in categories. Note that the same
        category_inputs not in categories will not be similar to each other either.
Raises:
    InvalidSchemaException: If a schema object does not have a corresponding node in the
        similarity space.

Initializes a new instance of the CategoricalSimilaritySpace.

This constructor sets up the space with the necessary configurations to encode and measure categorical
similarity based on the provided parameters.

Args:
    category_input (StringList | String | list[String | StringList]):
    The schema field containing input categories to be considered in the similarity space.
    Input contains one or more categories in a list if `StringList` is provided.
    If `String` is provided, then the input must be a single value.
    categories (list[str]): A list of all the recognized categories. Categories not included in this list will
        be treated as 'other', unless `uncategorized_as_category` is False.
    negative_filter (float, optional): A value used to represent unmatched categories in the encoding process.
        This allows for a penalizing non-matching categories - in contrast to them contributing 0 to similarity,
         it is possible to influence the similarity score negatively. Defaults to 0.0.
    uncategorized_as_category (bool, optional): Determines whether categories not listed in `categories` should
        be treated as a distinct 'other' category. Defaults to True.

Raises:
    InvalidSchemaException: If a schema object does not have a corresponding node in the similarity space,
    indicating a configuration or implementation error.

### Ancestors (in MRO)

* superlinked.framework.dsl.space.space.Space
* abc.ABC

### Instance variables

`annotation: str`
:

`category: superlinked.framework.dsl.space.space_field_set.SpaceFieldSet`
:

`uncategorized_as_category: bool`
: