Xiaozhong Liu

ORCID: 0000-0003-3477-8323
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Topic Modeling
  • Advanced Text Analysis Techniques
  • Natural Language Processing Techniques
  • Recommender Systems and Techniques
  • Complex Network Analysis Techniques
  • Advanced Graph Neural Networks
  • Expert finding and Q&A systems
  • Sentiment Analysis and Opinion Mining
  • Text and Document Classification Technologies
  • Semantic Web and Ontologies
  • Biomedical Text Mining and Ontologies
  • Information Retrieval and Search Behavior
  • scientometrics and bibliometrics research
  • Speech and dialogue systems
  • Open Education and E-Learning
  • Wikis in Education and Collaboration
  • Multimodal Machine Learning Applications
  • Misinformation and Its Impacts
  • Turfgrass Adaptation and Management
  • Web Data Mining and Analysis
  • Artificial Intelligence in Law
  • Spam and Phishing Detection
  • Domain Adaptation and Few-Shot Learning
  • Online Learning and Analytics
  • Scientific Computing and Data Management

Worcester Polytechnic Institute
2021-2025

Indiana University Bloomington
2015-2024

Indiana University
2012-2021

Alibaba Group (China)
2018-2020

Kent State University
2020

University of Alabama
2020

University of Cape Town
2020

Conference Board
2020

Association for Computing Machinery
2020

Alibaba Group (United States)
2018-2020

Understanding physiological and biochemical factors involved in heat‐stress injury would help improve heat tolerance of cool‐season grasses. The objective this study was to investigate lipid peroxidation cell membranes relation creeping bentgrass ( Agrostis palustris Huds.). Two cultivars differing tolerance, L‐93 (heat tolerant) Penncross sensitive) were grown under two temperature regimes: 22/16°C (day/night) 35/25°C for 56 d growth chambers. Photochemical efficiency (Fv/Fm) chlorophyll...

10.2135/cropsci2000.402503x article EN Crop Science 2000-03-01

10.1023/a:1015744126533 article EN Plant and Soil 2002-01-01

In this article, we use innovative full‐text citation analysis along with supervised topic modeling and network‐analysis algorithms to enhance classical bibliometric publication/author/venue ranking. By utilizing contexts extracted from a large number of publications, each or publication is represented by probability distribution over set predefined topics, where labeled an author‐contributed keyword. We then used publication/citation generate graph vertex prior edge transitioning...

10.1002/asi.22883 article EN Journal of the American Society for Information Science and Technology 2013-07-09

Rapid research progress in science and technology (S&T) continuously shifting workforce needs exert pressure on each other the educational training systems that link them. Higher education institutions aim to equip new generations of students with skills expertise relevant participation for decades come, but their offerings sometimes misalign commercial techniques forged at frontiers research. Here, we analyze visualize dynamic skill (mis-)alignment between academic push, industry pull,...

10.1073/pnas.1804247115 article EN Proceedings of the National Academy of Sciences 2018-12-10

The impact of elevated ultraviolet‐B radiation (UV‐B, 280–320 nm) on membrane systems and lipid peroxidation, possible involvement active oxygen radicals was investigated in leaves two UV‐B susceptible rice cultivars ( Oryza sativa L. cvs IR74 Dular). Rice seedlings were grown a greenhouse for 10 days then treated with biologically effective (UV‐B BE ) 28 days. Oxidative stress effects evaluated by measuring superoxide anion (O 2 generation rate, hydrogen peroxide (H O content,...

10.1111/j.1399-3054.1997.tb01000.x article EN Physiologia Plantarum 1997-10-01

The sheer volume of scholarly publications available online significantly challenges how scholars retrieve the new information and locate candidate reference papers. While classical text retrieval pseudo relevance feedback (PRF) algorithms can assist in accessing needed publications, this study, we propose an innovative publication ranking method with PRF by leveraging a number meta-paths on heterogeneous bibliographic graph. Different graph address different hypotheses, whereas...

10.1145/2661829.2661965 article EN 2014-11-03

In another study, results of growth responses to cytokinin indicate that heat stress injury in creeping bentgrass can be alleviated some extent by injection the root zone. The objective this study was examine whether alleviating effects exogenous application a on ( Agrostis palustris L.) involve regulation antioxidant activities. Creeping exposed three air/soil temperature regimes for 56 d chambers: (i) low air and soil control (20/20°C); (ii) high (20/35°C); (iii) temperatures (35/35°C)....

10.2135/cropsci2002.4660 article EN Crop Science 2002-03-01

While the volume of scholarly publications has increased at a frenetic pace, accessing and consuming useful candidate papers, in very large digital libraries, is becoming an essential challenging task for scholars. Unfortunately, because language barrier, some scientists (especially junior ones or graduate students who do not master other languages) cannot efficiently locate hosted foreign repository. In this study, we propose novel solution, cross-language citation recommendation via...

10.1145/3209978.3210032 article EN 2018-06-27

In a customer service system, dialogue summarization can boost efficiency by automatically creating summaries for long spoken dialogues in which customers and agents try to address issues about specific topics. this work, we focus on topic-oriented summarization, generates highly abstractive that preserve the main ideas from dialogues. dialogues, abundant noise common semantics could obscure underlying informative content, making general topic modeling approaches difficult apply. addition,...

10.1609/aaai.v35i16.17723 article EN Proceedings of the AAAI Conference on Artificial Intelligence 2021-05-18

Sentiment analysis in dialogues plays a critical role dialogue data analysis. However, previous studies on sentiment classification largely ignore topic information, which is important for capturing overall information some types of dialogues. In this study, we focus the task an type dialogue, namely customer service and propose novel approach captures to enhance performance. Specifically, topic-aware multi-task learning (TML) learns topic-enriched utterance representations by various kinds...

10.1609/aaai.v34i05.6454 article EN Proceedings of the AAAI Conference on Artificial Intelligence 2020-04-03

Retrieval-Augmented Generation (RAG) addresses hallucination and real-time constraints by dynamically retrieving relevant information from a knowledge database to supplement the LLMs' input. When presented with query, RAG selects most semantically similar texts its bases uses them as context for LLMs generate more accurate responses. also creates new attack surface, especially since databases are frequently sourced public domains. While existing studies have predominantly focused on...

10.48550/arxiv.2501.02968 preprint EN arXiv (Cornell University) 2025-01-06

The World Health Organization (WHO) declared the COVID-19 outbreak a Public Emergency of International Concern (PHEIC) on January 31, 2020. However, rumors "mysterious virus" had already been circulating in China December 2019, possibly preceding first confirmed case. Understanding how awareness about an emerging pandemic spreads through society is vital not only for enhancing disease surveillance, but also mitigating demand shocks and social inequities, such as shortages personal protective...

10.48550/arxiv.2502.05622 preprint EN arXiv (Cornell University) 2025-02-08

Abstract The World Health Organization (WHO) declared the COVID-19 outbreak a Public Emergency of International Concern on 2020 January 31. However, rumors “mysterious virus” had already been circulating in China 2019 December, possibly preceding first confirmed case. Understanding how awareness about an emerging pandemic spreads through society is vital not only for enhancing disease surveillance, but also mitigating demand shocks and social inequities, such as shortages personal protective...

10.1093/pnasnexus/pgaf043 article EN cc-by PNAS Nexus 2025-02-01

Heat stress injury may involve inhibition of cytokinin biosynthesis in roots. The objective this study was to examine whether application a synthetic cytokinin, zeatin riboside (ZR), the root zone would enhance tolerance creeping bentgrass (Agrostis palustris L.) high soil temperature or combination with air temperature. Grasses were exposed three and regimes for 56 d growth chambers: (i) optimum (control), 20/20°C; (ii) but (20/35°C); (iii) temperatures (35/35°C). Four concentrations (0.01,...

10.2135/cropsci2002.0457 article EN Crop Science 2002-01-01

Large volumes of textual data pose considerable challenges for manual qualitative analysis. We explore semi-automatic coding by leveraging Natural Language Processing (NLP). compare the performance human-developed NLP rules to those inferred machine learning (ML) algorithms. The results suggest that with ML may be useful support researchers data.

10.1002/meet.14504701328 article EN Proceedings of the American Society for Information Science and Technology 2010-11-01

In recent years social media has been used to characterize and predict real world events, in this research we seek investigate how closely Twitter mirrors the world. Specifically, wish relationship between language on results of 2011 NBA Playoff games. We hypothesize that by users will be useful classifying users' locations combined with current status which team is lead during game. This based common assumption "fans" a have more positive sentiment accordingly use different when their doing...

10.1145/2513549.2513559 article EN 2013-10-28

In the literature, existing studies on aspect sentiment classification (ASC) focus individual non-interactive reviews. This paper extends research to interactive reviews and proposes a new task, namely Aspect Sentiment Classification towards Question-Answering (ASC-QA), for real-world applications. task aims predict polarities specific aspects from QA style particular, high-quality annotated corpus is constructed ASC-QA facilitate corresponding research. On this basis, Reinforced...

10.18653/v1/p19-1345 article EN cc-by 2019-01-01
Coming Soon ...