Author name cluster

Shifeng Liu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

EAAI Journal 2022 Journal Article

A novel two-stage TOPSIS approach based on interval-valued probabilistic linguistic q-rung orthopair fuzzy sets with its application to MAGDM problems

Yuan Xu
Shifeng Liu
Jun Wang
Xiaopu Shang

Fuzzy theories are widely used in multi-attribute group decision-making (MAGDM) problems to describe uncertain and hesitant information. The recently proposed probabilistic linguistic q-rung orthopair fuzzy set (PLq-ROFS) is capable in dealing with quantitative and qualitative information simultaneously. However, in many actual situations, decision-makers (DMs) prefer to utilize interval values to express their minds and evaluations, this paper employs interval values to represent the probabilistic distribution of the membership degrees (MDs) and non-membership degrees (NMDs), and proposes the interval-valued PLq-ROFS (IVPLq-ROFS). Based on which, a novel two-stage TOPSIS approach is introduced. The IVPLq-ROFS weighted extended power average (IVPLq-ROFWEPA) operator is proposed to obtain the comprehensive matrix and weights of each attribute, while a further advanced TOPSIS model is constructed to get the final rank of alternatives. An example of a new-type smart city development evaluation problem is given to illustrate the efficacy of the proposed approach. Results show that our approach is more flexible, more adaptable, more accurate, more freedom, and has much lower calculation complexity in the calculation process of the MAGDM problem. The contributions of the proposed method are mainly manifested in giving the concept of IVPLq-ROFSs, and proposing a novel two-stage TOPSIS model for dealing with MAGDM problems under IVPLq-ROFSs.

Details DOI

AAAI Conference 2020 Conference Paper

HAMNER: Headword Amplified Multi-Span Distantly Supervised Method for Domain Specific Named Entity Recognition

Shifeng Liu
Yifang Sun
Bing Li
Wei Wang
Xiang Zhao

To tackle Named Entity Recognition (NER) tasks, supervised methods need to obtain sufﬁcient cleanly annotated data, which is labor and time consuming. On the contrary, distantly supervised methods acquire automatically annotated data using dictionaries to alleviate this requirement. Unfortunately, dictionaries hinder the effectiveness of distantly supervised methods for NER due to its limited coverage, especially in speciﬁc domains. In this paper, we aim at the limitations of the dictionary usage and mention boundary detection. We generalize the distant supervision by extending the dictionary with headword based non-exact matching. We apply a function to better weight the matched entity mentions. We propose a span-level model, which classiﬁes all the possible spans then infers the selected spans with a proposed dynamic programming algorithm. Experiments on all three benchmark datasets demonstrate that our method outperforms previous state-of-the-art distantly supervised methods.

PDF Details

AAAI Conference 2020 Conference Paper

Recursively Binary Modification Model for Nested Named Entity Recognition

Bing Li
Shifeng Liu
Yifang Sun
Wei Wang
Xiang Zhao

Recently, there has been an increasing interest in identifying named entities with nested structures. Existing models only make independent typing decisions on the entire entity span while ignoring strong modiﬁcation relations between subentity types. In this paper, we present a novel Recursively Binary Modiﬁcation model for nested named entity recognition. Our model utilizes the modiﬁcation relations among sub-entities types to infer the head component on top of a Bayesian framework and uses entity head as a strong evidence to determine the type of the entity span. The process is recursive, allowing lower-level entities to help better model those on the outer-level. To the best of our knowledge, our work is the ﬁrst effort that uses modiﬁcation relation in nested NER task. Extensive experiments on four benchmark datasets demonstrate that our model outperforms state-of-the-art models in nested NER tasks, and delivers competitive results with state-of-the-art models in ﬂat NER task, without relying on any extra annotations or NLP tools.

PDF Details