Recommendation systems are gaining increasing popularity in many application areas like e-commerce, movie and music recommendations, tourism, news, advertisement, stock markets, social networks etc. Conventional recommendation systems either use content based or collaborative filtering based approaches to model user preferences and give recommendations. These systems usually fail to consider evolving user preferences in different contextual situations. Context Aware Recommendation Systems take different contextual attributes into consideration and try to capture user preferences correctly. This survey focuses on the state-of-the art computational intelligence techniques trying to improve conventional design using contextual information. Further, these techniques are grouped into bio-inspired computing techniques and statistical computing techniques. The literature related to these techniques mentioning their ability to handle challenges faced by Context Aware Recommendation System are presented in this survey. The survey also talks about context inclusion strategies, classification of the contexts used in the literature reviewed, their impact on the problems faced by the recommendation systems, effective usage of these contexts, datasets used in the domain, future research scope in all the reviewed techniques and overall future research directions and challenges.

推荐系统在电子商务，电影和音乐推荐，旅游，新闻，广告，股票市场，社交网络等许多应用领域中越来越受欢迎。常规推荐系统使用基于内容或基于协作过滤的方法来建模用户偏好和提出建议。这些系统通常无法考虑不同上下文情况下不断发展的用户偏好。上下文感知推荐系统考虑了不同的上下文属性，并尝试正确捕获用户偏好。这项调查着重于最新的计算智能技术，这些技术试图使用上下文信息来改进常规设计。进一步，这些技术分为生物启发式计算技术和统计计算技术。本次调查介绍了与这些技术有关的文献，这些文献提到了它们应对上下文感知推荐系统所面临挑战的能力。该调查还讨论了上下文包含策略，所审查文献中所用上下文的分类，它们对推荐系统所面临问题的影响，这些上下文的有效使用，该领域中使用的数据集，所有已审查技术中的未来研究范围以及未来的总体研究方向和挑战。