Context-aware Service Recommendation based on Knowledge Graph Embedding

Abstract : As a class of context-aware systems, context-aware service recommendation aims to bind high-quality services to users while taking into account their context requirements, including invocation time, location, social profiles, connectivity, and so on. However, current CASR approaches are not scalable with the huge amount of service data (QoS and context information, users reviews and feedbacks). In addition, they lack a rich representation of contextual information as they adopt a simple matrix view. Moreover, current CASR approaches adopt the traditional user-service relation and they do not allow for multi-relational interactions between users and services in different contexts. To offer a scalable and context-sensitive service recommendation with great analysis and learning capabilities, we provide a rich and multi-relational representation of the CASR knowledge, based on the concept of knowledge graph. The constructed context-aware service knowledge graph (C-SKG) is, then, transformed into a low-dimentional vector space to facilitate its processing. For this purpose, we adopt Dilated Recurrent Neural Networks to propose a context-aware knowledge graph embedding, based on the principles of first-order and subgraph-aware proximity. Finally, a recommendation algorithm is defined to deliver the top-rated services according to the target user's context. Experiments have proved the accuracy and scalability of our CASR approach.

 EXISTING SYSTEM :

 ? We present a music recommendation scenario that we will use as a running example throughout the paper. ? Existing recommendation systems tend to recommend songs of a specific music genre based on the user’s profile; e.g., recommending rock and jazz for male users in their 40s. ? The graph data model provides flexibility and extensibility, making it easy to integrate data from multiple sources. We use rules to integrate the information coming from various sources and to explicate its semantics. ? Data from smartphone profile service can include users’ profile data as well as some basic contextual information. This data has a well defined structure and is relatively easy to add to the personal knowledge graph.

 DISADVANTAGE :

 ? Knowledge graphs can provide complementary information to overcome the problems faced by collaborative and content-based filtering approaches , since their recommendations are not linked to ratings; instead, they use domain knowledge. ? Collaborative filtering suffers mainly from two problems: sparsity of users’ data when there are few interactions between the user and the items, and cold-start problem (new user and new item). ? Content-based filtering (CBF) suffers from some limitations such as overspecialization, limited content analysis, serendipity, and new user problems. ? The underlying advantage of doing this is to overcome the problem of data sparsity.

 PROPOSED SYSTEM :

 • In this paper we propose an architecture for a recommendation system that can provide high-quality context aware recommendations on mobile devices without compromising privacy or incurring large storage and communication costs. • In contrast, the proposed on-device system can use available real-time data for this purpose. • One way to achieve this could be to use recently proposed semantic sequence labeling approaches for automatic slot filling that exploit statistical machine learning techniques such as long short-term memory (LSTM), convolutional neural networks (CNN), recurrent neural networks (RNN), and conditional random fields (CRF).

 ADVANTAGE :

 ? These systems commonly combine collaborative filtering with content-based filtering or collaborative filtering with any other recommendation approach. The goal of combination is to leverage each approach’s advantages and to improve the overall system performance. ? On the contrary, in general, classical recommendation methods based on feature vectors overlook such connections, which may result in suboptimal performance, especially when there is data sparsity. ? Concerning the performance of the analyzed works, in general, an improvement is observed compared to other recommendation methods taken as a baseline. ? Most recommendation models focus on achieving good performance from the point of view of accuracy.