Code search is an important approach to improve effectiveness and efficiency of software development. The current studies commonly search target code based on either semantic or statistical information in large datasets. Semantic and statistical information have hidden relationships between them since they describe code snippets from different perspectives. In this work, we propose a joint embedding model of semantic and statistical features to improve the effectiveness of code annotation. Then, we implement a code search engine, i.e., JessCS, based on the joint embedding model. We evaluate JessCS on more than 1 million lines of code snippets and corresponding descriptions. The experimental results show that JessCS performs more effective than UNIF-based approach, with at least 13% improvements on the studied metrics.

How to arrange sales staff to visit offline stores reasonably is a critical task in the Fast Moving Consumer Goods (FMCG) industry. Based on the K-means and Dijkstra algorithms (K-D fusion algorithm), this paper proposed an algorithm to automatically allocate offline stores for sales staff and optimize the visiting path, thereby improving management efficiency. A new initial cluster center selection approach was proposed for the K-means algorithm to select its initial clustering center with the consideration of outlier points. The sales staff’s visiting path to the store was planned by the Dijkstra algorithm. The performance of our K-D fusion algorithm was evaluated in terms of grouping rationality and path planning optimization. Experimental results show that our algorithm can comprehensively consider several factors such as the number of stores, store types, and geographical locations, and distribute the workload more evenly to all the sales staff. In addition, it can also optimize the visiting path for sales staff, which can effectively improve the efficiency of sales staff to visit offline stores.

No abstract