Sequential patent trading recommendation using knowledge-aware attentional bidirectional long short-term memory network (KBiLSTM)

Abstract

With the rapid development of the patent marketplace, patent trading recommendation is required to mitigate the technology searching cost of patent buyers. Current research focuses on the recommendation based on existing patents of a company; a few studies take into account the sequential pattern of patent acquisition activities and the possible diversity of a company’s business interests. Moreover, the profiling of patents based on solely patent documents fails to capture the high-order information of patents. To bridge the gap, we propose a knowledge-aware attentional bidirectional long short-term memory network (KBiLSTM) method for patent trading recommendation. KBiLSTM uses knowledge graph embeddings to profile patents with rich patent information. It introduces bidirectional long short-term memory network (BiLSTM) to capture the sequential pattern in a company’s historical records. In addition, to address a company’s diverse technology interests, we design an attention mechanism to aggregate the company’s historical patents given a candidate patent. Experimental results on the United States Patent and Trademark Office (USPTO) data set show that KBiLSTM outperforms state-of-the-art baselines for patent trading recommendation in terms of F1 and normalised discounted cumulative gain (nDCG). The attention visualisation of randomly selected company intuitively demonstrates the recommendation effectiveness.

Interpretable patent recommendation with knowledge graph and deep learning

“…The application contexts mainly contain query-driven patent search and patent transfer contexts 8 . The former provides specific user needs like keywords or patent documents while the latter requires the mining of company needs 9 , 22 . Patents have heterogeneous information that can be used for recommendation, including texts (e.g., patent titles and abstracts), categories (e.g., the International Patent Classification), interactions (e.g., patent searching and assignment behavior of users), citations, and inventors of patents 23 .…”

Knowledge Graph-aware Deep Interest Extraction Network on Sequential Recommendation

“…Consequently, incorporation of auxiliary information into recommendations to mine more available information could achieve performance gains in simple networks. To solve the cold-start and matrix sparsity problems, different types of auxiliary information have been applied in recommendation fields [16], including profile information [17,18], social networks [19,20], review text [21], image and video information [22,23], and knowledge graph (KG) [24][25][26], etc.…”