Systems and methods for determining entity characteristics

1 . A computer-implemented method for determining entity characteristics, the method comprising:

determining, by the entity characterisation system, a numerical representation of each of a plurality of second entities, each numerical representation comprising a multi-dimensional vector characteristic of the respective second entity, wherein each second entity is associated with one or more first entities, and wherein the one or more first entities are service providers and the plurality of second entities are clients of the one or more first entities;

reducing, by the entity characterisation system, the dimensionality of the numerical representations of the plurality of second entities to produce a plurality of reduced dimensionality numerical representations;

wherein reducing the dimensionality of the numerical representations of the second entities comprises:

performing principal component analysis on the numerical representations of the second entities to generate a plurality of intermediate numerical representations; and

performing Uniform Manifold Approximation and Projection (UMAP) on the intermediate second entity numerical representations to generate the plurality of reduced dimensionality numerical representations;

determining, by the entity characterisation system, a specialisation of each of the one or more first entities based on the respective plurality of reduced dimensionality numerical representations, wherein the specialisation of each of the one or more first entities is characteristic of similarities between the second entities of the respective first entity, and wherein determining the specialisation of each of the one or more first entities comprises:

centring a kernel function on each of the plurality of respective reduced dimensionality numerical representations of the respective one or more first entities; and

summing each of the kernel functions to determine a specialisation function of the respective one or more first entities, wherein the specialisation function is indicative of the specialisation;

encoding, by the entity characterisation system, the specialisation function into one or more visual representation display objects; and

providing, by the entity characterisation system, the one or more visual representation display objects on a user interface of a first computing device, wherein the visual representation display objects provide a visual representation of the specialisation function, and wherein the visual representation comprises an indication of one or more areas of relatively high density indicative of one or more respective more common classes of client of the service provider and one or more areas of relatively low density indicative of less common classes of client of the service provider.

2 . The computer-implemented method of claim 1 ,

wherein the specialisation numerical representation comprises one or more local maximum of the specialisation function.

3 . The computer-implemented method of claim 1 , wherein determining the specialisation of each of the plurality of first entities comprises a kernel density estimation.

4 . The method of claim 1 , further comprising:

receiving, from a user of a second computing device, a candidate second entity interested in engaging services of a first entity that specialises in providing services to second entities operating in the industry of the candidate second entity;

determining, by the entity characterisation system, a numerical representation of the candidate second entity, wherein the numerical representation of the candidate second entity is a multi-dimensional vector characteristic of the candidate second entity;

reducing, by the entity characterisation system, the dimensionality of the numerical representation of the candidate second entity to produce a reduced dimensionality numerical representation of the candidate second entity;

comparing, by the entity characterisation system, the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the one or more first entities;

determining one or more suggested first entities specialising in providing services to second entities operating in the industry of the candidate second entity based on the proximity of the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the one or more first entities; and

providing, by the entity characterisation system, a recommendation display object of the one or more suggested first entities to the second computing device.

5 . The computer-implemented method of claim 4 , further comprising classifying the candidate second entity, wherein said classifying comprises determining a matching score for one or more of the first entities, the matching score being indicative of a suitability of the relevant first entity to the candidate second entity.

6 . The computer-implemented method of claim 5 , wherein the matching score is based on the comparison of the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the plurality of first entities.

7 . The computer-implemented method of claim 6 , the matching scores are each related to a distance between the reduced dimensionality numerical representation of the candidate second entity and the relevant specialisation numerical representation.

8 . The computer-implemented method of claim 5 , wherein the one or more suggested first entities are determined based on the matching scores.

9 . A non-transitory computer-readable storage medium storing instructions that, when executed by a computer, cause the computer to perform operations including:

determining a numerical representation of each of a plurality of second entities, each numerical representation comprising a multi-dimensional vector characteristic of the respective second entity, wherein each second entity is associated with one or more first entities, and wherein the one or more first entities are service providers and the plurality of second entities are clients of the one or more first entities;

reducing the dimensionality of the numerical representations of the plurality of second entities to produce a plurality of reduced dimensionality numerical representations; wherein reducing the dimensionality of the numerical representations of the second entities comprises:

performing principal component analysis on the numerical representations of the second entities to generate a plurality of intermediate numerical representations; and

performing Uniform Manifold Approximation and Projection (UMAP) on the intermediate second entity numerical representations to generate the plurality of reduced dimensionality numerical representations;

determining a specialisation of each of the plurality of first entities based on the respective plurality of reduced dimensionality numerical representations, wherein the specialisation of each of the one or more first entities is characteristic of similarities between the second entities of the respective first entity, and wherein determining the specialisation of each of the one or more first entities comprises:

centring a kernel function on each of the plurality of respective reduced dimensionality numerical representations of the respective one or more first entities; and

summing each of the kernel functions to determine a specialisation function of the respective one or more first entities, wherein the specialisation function is indicative of the specialisation;

encoding, by the entity characterisation system, the specialisation function into one or more visual representation display objects; and providing, by the entity characterisation system, the one or more visual representation display objects on a user interface of a first computing device, wherein the visual representation display objects provide a visual representation of the specialisation function, and wherein the visual representation comprises an indication of one or more areas of relatively high density indicative of one or more respective more common classes of client of the service provider and one or more areas of relatively low density indicative of less common classes of client of the service provider.

10 . The non-transitory computer-readable storage medium of claim 9 , wherein the operations further include:

receiving, from a user of a second computing device, a candidate second entity interested in engaging services of a first entity that specialises in providing services to second entities operating in the industry of the candidate second entity;

determining, by the entity characterisation system, a numerical representation of the candidate second entity, wherein the numerical representation of the candidate second entity is a multi-dimensional vector characteristic of the candidate second entity;

reducing, by the entity characterisation system, the dimensionality of the numerical representation of the candidate second entity to produce a reduced dimensionality numerical representation of the candidate second entity;

comparing, by the entity characterisation system, the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the one or more first entities;

determining one or more suggested first entities specialising in providing services to second entities operating in the industry of the candidate second entity based on the proximity of the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the one or more first entities; and

providing, by the entity characterisation system, a recommendation display object of the one or more suggested first entities to the second computing device.

11 . The non-transitory computer-readable storage medium of claim 10 , wherein the operations further include classifying the candidate second entity, wherein said classifying comprises determining a matching score for one or more of the first entities, the matching score being indicative of a suitability of the relevant first entity to the candidate second entity.

12 . The non-transitory computer-readable storage medium of claim 11 , wherein the matching score is based on the comparison of the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the plurality of first entities.

13 . The non-transitory computer-readable storage medium of claim 12 , wherein the matching scores are each related to a distance between the reduced dimensionality numerical representation of the candidate second entity and the relevant specialisation numerical representation.

14 . The computer-implemented method of claim 11 , wherein the one or more suggested first entities are determined based on the matching scores.

15 . An entity characterisation system comprising:

at least one processor; and

a memory storing program instructions accessible by the at least one processor, and configured to cause the at least one processor to:

determine a numerical representation of each of a plurality of second entities, each numerical representation comprising a multi-dimensional vector characteristic of the respective second entity, wherein each second entity is associated with one or more first entities, and wherein the one or more first entities are service providers and the plurality of second entities are clients of the one or more first entities;

reduce the dimensionality of the numerical representations of the plurality of second entities to produce a plurality of reduced dimensionality numerical representations; wherein reducing the dimensionality of the numerical representations of the second entities comprises:

performing principal component analysis on the numerical representations of the second entities to generate a plurality of intermediate numerical representations; and

performing Uniform Manifold Approximation and Projection (UMAP) on the intermediate second entity numerical representations to generate the plurality of reduced dimensionality numerical representations;

determine a specialisation of each of the plurality of first entities based on the respective plurality of reduced dimensionality numerical representations, wherein the specialisation of each of the one or more first entities is characteristic of similarities between the second entities of the respective first entity, and wherein determining the specialisation of each of the one or more first entities comprises:

centring a kernel function on each of the plurality of respective reduced dimensionality numerical representations of the respective one or more first entities; and

summing each of the kernel functions to determine a specialisation function of the respective one or more first entities, wherein the specialisation function is indicative of the specialisation; encoding, by the entity characterisation system, the specialisation function into one or more visual representation display objects; and

providing, by the entity characterisation system, the one or more visual representation display objects on a user interface of a first computing device, wherein the visual representation display objects provide a visual representation of the specialisation function, and wherein the visual representation comprises an indication of one or more areas of relatively high density indicative of one or more respective more common classes of client of the service provider and one or more areas of relatively low density indicative of less common classes of client of the service provider.

16 . The entity characterisation system of claim 15 , further configured to:

receive, from a user of a second computing device, a candidate second entity interested in engaging services of a first entity that specialises in providing services to second entities operating in the industry of the candidate second entity;

determine, by the entity characterisation system, a numerical representation of the candidate second entity, wherein the numerical representation of the candidate second entity is a multi-dimensional vector characteristic of the candidate second entity;

reduce, by the entity characterisation system, the dimensionality of the numerical representation of the candidate second entity to produce a reduced dimensionality numerical representation of the candidate second entity;

compare, by the entity characterisation system, the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the one or more first entities;

determine one or more suggested first entities specialising in providing services to second entities operating in the industry of the candidate second entity based on the proximity of the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the one or more first entities; and

provide, by the entity characterisation system, a recommendation display object of the one or more suggested first entities to the second computing device.

17 . The entity characterisation system of claim 16 , further configured to classify the candidate second entity, wherein said classifying comprises determining a matching score for one or more of the first entities, the matching score being indicative of a suitability of the relevant first entity to the candidate second entity.

18 . The entity characterisation system of claim 17 , wherein the matching score is based on the comparison of the reduced dimensionality numerical representation of the candidate second entity to the specialisation of each of the plurality of first entities.

19 . The entity characterisation system of claim 18 , wherein the matching scores are each related to a distance between the reduced dimensionality numerical representation of the candidate second entity and the relevant specialisation numerical representation.

20 . The entity characterisation system of claim 17 , wherein the one or more suggested first entities are determined based on the matching scores.