Recommender system

A recommender system is usually a computer system which makes suggestions to the user about how to do a task.

What they do is make suggestions rather than actually solve problems. One key objective is to make sure the user has considered standard issues in the domain of choice. By domain is meant subject or topic. It is a weakness of regular AI software that it is not good for problems which involve personal preference or choice. The key concept for a recommender system is "relevance".

It is already clear that many data bases have so much material that manual searches are bound to miss some important items. The first actual mention of the term "recommender system" was in a technical report as a "digital bookshelf" in 1990 by Jussi Karlgren at Columbia University.^[1]

References[change | change source]

↑ Karlgren, Jussi. 1990. "An Algebra for Recommendations". Syslab Working Paper 179 (1990)

[1] Karlgren, Jussi. 1990. "An Algebra for Recommendations". Syslab Working Paper 179 (1990)

[1]

v t e Machine learning evaluation metrics
Regression	MSE MAE sMAPE MAPE MASE MSPE RMS RMSE/RMSD R² MDA MAD
Classification	F-score P4 Accuracy Precision Recall Kappa MCC AUC ROC Sensitivity and specificity Logarithmic Loss
Clustering	Silhouette Calinski-Harabasz Davies-Bouldin Dunn index Hopkins statistic Jaccard index Rand index Similarity measure SMC SimHash
Ranking	MRR DCG NDCG AP
Computer Vision	PSNR SSIM IoU
NLP	Perplexity BLEU
Deep Learning Related Metrics	Inception score FID
Recommender system	Coverage Intra-list Similarity
Similarity	Cosine similarity Euclidean distance Pearson correlation coefficient
Confusion matrix