Lei Fang

Lei Fang

Auto-generated profile, based on publicly available information

Lei Fang

Microsoft Research Asia, Beijing, China

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 August 2019 in IEEE Transactions on Visualization and Computer Graphics
Verified
Journal Article
Closed access
WCWeiwei Cui
Weiwei Cui
XZXiaoyu Zhang
Xiaoyu Zhang
YWYun Wang
Yun Wang
HHHe Huang
He Huang
BCBei Chen
Bei Chen
DZDongmei Zhang
Dongmei Zhang
Combining data content with visual embellishments, infographics can effectively deliver messages in an engaging and memorable manner. Various authoring tools have been proposed to facilitate the creation of infographics. However, creating a professional infographic with these authoring tools is still not an easy task, requiring much time and design expertise. Therefore, these tools are generally not attractive to casual users, who are either unwilling to take time to learn the tools or lacking in proper design expertise to create a professional infographic. In this paper, we explore an alternative approach: to automatically generate infographics from natural language statements. We first conducted a preliminary study to explore the design space of infographics. Based on the preliminary study, we built a proof-of-concept system that automatically converts statements about simple proportionrelated statistics to a set of infographics with pre-designed styles. Finally, we demonstrated the usability and usefulness of the system through sample results, exhibits, and expert reviews.

28 April 2023 in Nature Communications
Verified
Journal Article
Open access
PDF file
LFLei Fang
Lei Fang
JLJunren Li
Junren Li
MZMing Zhao
Ming Zhao
LTLi Tan
Li Tan
JLJian-Guang Lou
Jian-Guang Lou
Retrosynthesis analysis is an important task in organic chemistry with numerous industrial applications. Previously, machine learning approaches employing natural language processing techniques achieved promising results in this task by first representing reactant molecules as strings and subsequently predicting reactant molecules using text generation or machine translation models. Chemists cannot readily derive useful insights from traditional approaches that rely largely on atom-level decoding in the string representations, because human experts tend to interpret reactions by analyzing substructures that comprise a molecule. It is well-established that some substructures are stable and remain unchanged in reactions. In this paper, we developed a substructure-level decoding model, where commonly preserved portions of product molecules were automatically extracted with a fully data-driven approach. Our model achieves improvement over previously reported models, and we demonstrate that its performance can be boosted further by enhancing the accuracy of these substructures. Analyzing substructures extracted from our machine learning model can provide human experts with additional insights to assist decision-making in retrosynthesis analysis.

2 February 2024 in Digital Discovery
Verified
Journal Article
Open access
PDF file
JLJunren Li
Junren Li
LFLei Fang
Lei Fang
JLJian-Guang Lou
Jian-Guang Lou
Retro-BLEU is a statistical metric to evaluate the plausibility of model-generated retrosynthesis routes based on reaction template sequences analysis.

8 June 2023 in Journal of Cheminformatics
Verified
Journal Article
Open access
PDF file
JLJunren Li
Junren Li
LFLei Fang
Lei Fang
JLJian-Guang Lou
Jian-Guang Lou
Retrosynthesis is an important task in organic chemistry. Recently, numerous data-driven approaches have achieved promising results in this task. However, in practice, these data-driven methods might lead to sub-optimal outcomes by making predictions based on the training data distribution, a phenomenon we refer as frequency bias. For example, in template-based approaches, low-ranked predictions are typically generated by less common templates with low confidence scores which might be too low to be comparable, and it is observed that recorded reactants can be among these low-ranked predictions. In this work, we introduce RetroRanker, a ranking model built upon graph neural networks, designed to mitigate the frequency bias in predictions of existing retrosynthesis models through re-ranking. RetroRanker incorporates potential reaction changes of each set of predicted reactants in obtaining the given product to lower the rank of chemically unreasonable predictions. The predicted re-ranked results on publicly available retrosynthesis benchmarks demonstrate that we can achieve improvement on most state-of-the-art models with RetroRanker. Our preliminary studies also indicate that RetroRanker can enhance the performance of multi-step retrosynthesis.

1 August 2020
Verified
Conference Paper
Closed access
MZMingzhu Zhang
Mingzhu Zhang
YZYang Zhongguo
Yang Zhongguo
CLChen Liu
Chen Liu
LFLei Fang
Lei Fang
Most of Traditional Chinese Medicine (TCM) data and ancient records exist in the form of books. The unstructured medical information is the foundation for building TCM knowledge service. The existing methods are not accurate enough to solve TCM named entity recognition and require a lot of manual labeling data. This paper proposes a semi-supervised embedded Semi-BERT-BiLSTM-CRF model. Based on the book “Diagnosis of Traditional Chinese Medicine in Traditional Chinese Medicine”, we select the physical features from the cleaned-up text information according to the characteristics of Chinese medicine classics, and then use a small amount of labeled data to train the BERT-BiLSTM-CRF model. The obtained model is used to predict unlabeled data and obtain pseudo-label data. The pseudo-label and labeled data are used as a training set for model training. Experiments show that TCM entity recognition accuracy of this method reaches 81.24%, which effectively improves the TCM entity recognition accuracy and reduces the manual labeling work. The results of this research can be applied to scenarios such as auxiliary diagnosis of TCM and expert system after subsequent improvement and transformation.

1 October 2011 in International Journal of Applied Evolutionary Computation
Verified
Journal Article
Closed access
PDF file
LFLei Fang
Lei Fang
Steven GuanSteven Guan
Steven Guan
HZHaofan Zhang
Haofan Zhang
Rule-based Genetic Algorithms (GAs) have been used in the application of pattern classification (Corcoran & Sen, 1994), but conventional GAs have weaknesses. First, the time spent on learning is long. Moreover, the classification accuracy achieved by a GA is not satisfactory. These drawbacks are due to existing undesirable features embedded in conventional GAs. The number of rules within the chromosome of a GA classifier is usually set and fixed before training and is not problem-dependent. Secondly, conventional approaches train the data in batch without considering whether decomposition solves the problem. Thirdly, when facing large-scale real-world problems, GAs cannot utilise resources efficiently, leading to premature convergence. Based on these observations, this paper develops a novel algorithmic framework that features automatic domain and task decomposition and problem-dependent chromosome length (rule number) selection to resolve these undesirable features. The proposed Recursive Learning of Genetic Algorithm with Task Decomposition and Varied Rule Set (RLGA) method is recursive and trains and evolves a team of learners using the concept of local fitness to decompose the original problem into sub-problems. RLGA performs better than GAs and other related solutions regarding training duration and generalization accuracy according to the experimental results.