Add Intra-List Diversity (ILD) Metric

[nikitakumari2]

Problem:
The rexmex library offers a strong set of metrics for evaluating recommender systems, with a great focus on fairness and coverage. However, it could be enhanced by including metrics that specifically measure the diversity of recommendations. A common scenario is when a recommender is accurate but repeatedly suggests very similar items, leading to a poor user experience and "filter bubble" effects. Measuring diversity is crucial for evaluating how well a model avoids this behavior.

Proposed Solution:
I propose implementing Intra-List Diversity (ILD), a standard "beyond accuracy" metric that measures the average dissimilarity between all pairs of items within a recommendation list. A higher ILD score indicates a more diverse set of recommendations.

Academic References:
This metric is well established in the recommender systems literature. Key references include:

1. Vargas, S., & Castells, P. (2011). Novelty and Diversity in Recommender Systems: Choice, Discovery and Relevance. In Proceedings of the 5th ACM conference on Recommender systems (RecSys '11).

2. Carbonell, J., & Goldstein, J. (1998). The Use of MMR, Diversity-Based Reranking for Reordering Documents and Producing Summaries. In Proceedings of the 21st annual international ACM SIGIR conference.