问题描述
我正在尝试为少于1000个用户和大约2000个项目训练火柴盒"推荐.通常结果很差，因为仅向所有用户推荐了一些项目.此外，我看不到用户功能或项目功能有任何作用 根据建议.我尝试了多种特征，迭代和培训批次.
I am trying to train a Matchbox recommendation for less than 1000 users and about 2000 items. The results are bad in general as only a few items are recommended to all the users. Furthermore, I don't see the user features or item features have any effect on the recommendation. I have tried multiple numbers of traits, iterations and training batches.
请让我知道如何改善结果.
Kindly let me know how I can improve the result.
谢谢
推荐答案
很抱歉听到您在使用Matchbox推荐器时遇到问题，因为我没有有关您的实验的详细信息，所以我无法分辨出问题所在.但我建议您参考我们的示例实验:https://gallery.azure.ai/Experiment/3a02931f94114f47b4512dd9179b515e
Sorry to hear you have problem when using Matchbox recommender, since I don't have the details about your experiment, I can't tell what the problem is. But I would recommend you to refer to our sample experiment: https://gallery.azure.ai/Experiment/3a02931f94114f47b4512dd9179b515e
此外，如果您想了解有关Matchbox推荐模块的更多详细信息，请参阅:https://docs.microsoft.com/zh-cn/azure/machine-learning/studio-module-reference/score-火柴盒推荐
And also, if you want to know more details about Matchbox Recommender module, please refer to: https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/score-matchbox-recommender
此致
雨桐
这篇关于用户功能和项目功能对Matchbox建议不起作用的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！