Text2SQL can help non-professionals connect with databases by turning natural languages into SQL. Although previous researches about Text2SQL have provided some workable solutions, most of them extract values based on column representation. If there are multiple values in the query and these values belong to different columns, the previous approaches based on column representation cannot accurately extract values. In this work, we propose a new neural network architecture based on the pre-trained BERT, called M-SQL. The column-based value extraction is divided into two modules, value extraction and value-column matching. We evaluate M-SQL on a more complicated TableQA dataset, which comes from an AI competition. We rank first in this competition. Experimental results and competition ranking show that our proposed M-SQL achieves state-of-the-art results on TableQA.

Person re-identification (Re-ID) has become a hot topic in both research and industry. We joined in a person Re-ID challenge of the First National Artificial Intelligence Challenge (China, 2019) and found some model designs and training tricks work great or not on a super big private dataset. In this paper, we propose a model that combines the most effective designs, including multi-scale, multi-branch and attention mechanism, and report training tricks that are no less or even more important in improving person Re-ID performance. We analyze four commonly used public datasets: Market1501, DukeMTMC-ReID, CUHK03, and MSMT17, and achieve the state-of-the-art performance. Besides, we analyze and confirm the effectiveness of the designs by ablation studies. We also share strategies that play a key role in the challenge and experience of model designs that do not generalize well on large datasets.

Automatically parsing SQL queries from natural languages can help non-professionals access databases and improve the efficiency of information utilization. It is a long-term research issue and has recently received attention from the relevant communities. Although previous researches have provided some workable solutions, most of them only consider table schemas and natural language questions when parsing SQL queries, and do not use table contents. We observe that table contents can provide more helpful information for some user questions. In this paper, we propose a novel neural network approach, F-SQL, to focus on solving the problem of table content utilization. In particular, we employ the gate mechanism to fuse table schemas and table contents and get the more different representation about table schemas. We test this idea on the WikiSQL and TableQA datasets. Experimental results show that F-SQL achieves new state-of-the-art results on WikiSQL and TableQA.

As edge computing has found its way into people's everyday life, many app vendors put the quality of experience (QoE) in an important position. QoE experienced by users of different application services vary even with same resources. Because of the restricted resource of edge computing nodes, it's very necessary to assign different amounts of resources to different kinds of services to reach better QoE. In this paper, we need to decide how many edge node resources each user can get respectively in the service area. When allocating resources, app vendors should fully consider the fairness of resource allocation and maximize the total users' experience. We first establish the Multi-service Q oE model for four popular services. We determine the multi-objective function to evaluate fairness and utility by setting the resource impact factors of different service types. Being an NP-hard problem, the classical NSGA-II algorithm combined with CPLEX solver is proposed to accelerate the process of solution. Then experiments are carried out on simulated data, and the applicability of the algorithm is evaluated in user-sparse scenarios, user-intensive scenarios and other scenarios.

Traditional collaborative filtering recommendation technology has lower accuracy and couldn't meet the actual needs. So we convert the recommendation problem into the classification problem. We introduce Logistic classification methods based on the collaborative filtering technology and determine whether the recommending units are recommendable according to personal feature of users and products. The result shows that the method could greatly improve the accuracy of recommendation under the premise to ensure recall.