Zhongwei Xie
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Zhongwei Xie has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
Lin Li
School of Computer Science and Artificial Intelligence, Wuhan University of Technology, Wuhan, Hubei, China
3 shared publications- Lin Li
Wuhan University of Technology, Wuhan, China
3 shared publications - Ling Liu
School of Computer Science, Georgia Institute of Technology
2 shared publications Luo Zhong
School of Computer Science and Artificial Intelligence, Wuhan University of Technology, Wuhan, Hubei, China
2 shared publications- Yanzhao Wu
Florida International University, Miami, FL, USA
2 shared publications - Haohan Zhang
Wuhan University of Technology
1 shared publication
Latest Publications
- Open Access
- 29 July 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 15 (4) , 1-28
Boosting Healthiness Exposure in Category-Constrained Meal Recommendation Using Nutritional Standards
Food computing, a newly emerging topic, is closely linked to human life through computational methodologies. Meal recommendation, a food-related study about human health, aims to provide users a meal with courses constrained from specific categories (e.g., appetizers, main dishes) that can be enjoyed as a service. Historical interaction data, important user information, is often used by existing models to learn user preferences. However, if a user’s preferences favor less healthy meals, the model will follow that preference and make similar recommendations, potentially negatively impacting the user’s long-term health. This emphasizes the necessity for health-oriented and responsible meal recommendation systems. In this article, we propose a healthiness-aware and category-wise meal recommendation model called CateRec, which boosts healthiness exposure by using nutritional standards as knowledge to guide the model training. Two fundamental questions are raised and answered: (1) How can the healthiness of meals be evaluated? Two well-known nutritional standards from the World Health Organization and the United Kingdom Food Standards Agency are used to calculate the healthiness score of the meal. (2) How can the model training be guided in a health-oriented manner? We construct category-wise personalization partial rankings and category-wise healthiness partial rankings, and theoretically analyze that they meet the necessary properties and assumptions required to be trained by the maximum posterior estimator under Bayesian probability. The data analysis confirms the existence of user preferences leaning towards less healthy meals in two public datasets. A comprehensive experiment demonstrates that our CateRec effectively boosts healthiness exposure in terms of mean healthiness score and ranking exposure while being comparable to the state-of-the-art model in terms of recommendation accuracy.
- 29 July 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 15 (4) , 1-28
Open AccessBoosting Healthiness Exposure in Category-Constrained Meal Recommendation Using Nutritional Standards
Food computing, a newly emerging topic, is closely linked to human life through computational methodologies. Meal recommendation, a food-related study about human health, aims to provide users a meal with courses constrained from specific categories (e.g., appetizers, main dishes) that can be enjoyed as a service. Historical interaction data, important user information, is often used by existing models to learn user preferences. However, if a user’s preferences favor less healthy meals, the model will follow that preference and make similar recommendations, potentially negatively impacting the user’s long-term health. This emphasizes the necessity for health-oriented and responsible meal recommendation systems. In this article, we propose a healthiness-aware and category-wise meal recommendation model called CateRec, which boosts healthiness exposure by using nutritional standards as knowledge to guide the model training. Two fundamental questions are raised and answered: (1) How can the healthiness of meals be evaluated? Two well-known nutritional standards from the World Health Organization and the United Kingdom Food Standards Agency are used to calculate the healthiness score of the meal. (2) How can the model training be guided in a health-oriented manner? We construct category-wise personalization partial rankings and category-wise healthiness partial rankings, and theoretically analyze that they meet the necessary properties and assumptions required to be trained by the maximum posterior estimator under Bayesian probability. The data analysis confirms the existence of user preferences leaning towards less healthy meals in two public datasets. A comprehensive experiment demonstrates that our CateRec effectively boosts healthiness exposure in terms of mean healthiness score and ranking exposure while being comparable to the state-of-the-art model in terms of recommendation accuracy.
- 14 November 2023
- Published by Springer Nature
- p. 452-467
Transductive Cross-Lingual Scene-Text Visual Question Answering
- 14 November 2023
- Published by Springer Nature
- p. 452-467
Transductive Cross-Lingual Scene-Text Visual Question Answering
- Open Access
- 1 December 2021
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 40 (4) , 1-27
Learning Text-image Joint Embedding for Efficient Cross-modal Retrieval with Deep Feature Engineering
This article introduces a two-phase deep feature engineering framework for efficient learning of semantics enhanced joint embedding, which clearly separates the deep feature engineering in data preprocessing from training the text-image joint embedding model. We use the Recipe1M dataset for the technical description and empirical validation. In preprocessing, we perform deep feature engineering by combining deep feature engineering with semantic context features derived from raw text-image input data. We leverage LSTM to identify key terms, deep NLP models from the BERT family, TextRank, or TF-IDF to produce ranking scores for key terms before generating the vector representation for each key term by using Word2vec. We leverage Wide ResNet50 and Word2vec to extract and encode the image category semantics of food images to help semantic alignment of the learned recipe and image embeddings in the joint latent space. In joint embedding learning, we perform deep feature engineering by optimizing the batch-hard triplet loss function with soft-margin and double negative sampling, taking into account also the category-based alignment loss and discriminator-based alignment loss. Extensive experiments demonstrate that our SEJE approach with deep feature engineering significantly outperforms the state-of-the-art approaches.
- 1 December 2021
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 40 (4) , 1-27
Open AccessLearning Text-image Joint Embedding for Efficient Cross-modal Retrieval with Deep Feature Engineering
This article introduces a two-phase deep feature engineering framework for efficient learning of semantics enhanced joint embedding, which clearly separates the deep feature engineering in data preprocessing from training the text-image joint embedding model. We use the Recipe1M dataset for the technical description and empirical validation. In preprocessing, we perform deep feature engineering by combining deep feature engineering with semantic context features derived from raw text-image input data. We leverage LSTM to identify key terms, deep NLP models from the BERT family, TextRank, or TF-IDF to produce ranking scores for key terms before generating the vector representation for each key term by using Word2vec. We leverage Wide ResNet50 and Word2vec to extract and encode the image category semantics of food images to help semantic alignment of the learned recipe and image embeddings in the joint latent space. In joint embedding learning, we perform deep feature engineering by optimizing the batch-hard triplet loss function with soft-margin and double negative sampling, taking into account also the category-based alignment loss and discriminator-based alignment loss. Extensive experiments demonstrate that our SEJE approach with deep feature engineering significantly outperforms the state-of-the-art approaches.
- 26 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 15 (6) , 3304-3316
Learning TFIDF Enhanced Joint Embedding for Recipe-Image Cross-Modal Retrieval Service
It is widely acknowledged that learning joint embeddings of recipes with images is challenging due to the diverse composition and deformation of ingredients in cooking procedures. We present a Multi-modal Semantics enhanced Joint Embedding approach (MSJE) for learning a common feature space between the two modalities (text and image), with the ultimate goal of providing high-performance cross-modal retrieval services. Our MSJE approach has three unique features. First, we extract the TFIDF feature from the title, ingredients and cooking instructions of recipes. By determining the significance of word sequences through combining LSTM learned features with their TFIDF features, we encode a recipe into a TFIDF weighted vector for capturing significant key terms and how such key terms are used in the corresponding cooking instructions. Second, we combine the recipe TFIDF feature with the recipe sequence feature extracted through two-stage LSTM networks, which is effective in capturing the unique relationship between a recipe and its associated image(s). Third, we further incorporate TFIDF enhanced category semantics to improve the mapping of image modality and to regulate the similarity loss function during the iterative learning of cross-modal joint embedding. Experiments on the benchmark dataset Recipe1M show the proposed approach outperforms the state-of-the-art approaches.
- 26 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 15 (6) , 3304-3316
Learning TFIDF Enhanced Joint Embedding for Recipe-Image Cross-Modal Retrieval Service
It is widely acknowledged that learning joint embeddings of recipes with images is challenging due to the diverse composition and deformation of ingredients in cooking procedures. We present a Multi-modal Semantics enhanced Joint Embedding approach (MSJE) for learning a common feature space between the two modalities (text and image), with the ultimate goal of providing high-performance cross-modal retrieval services. Our MSJE approach has three unique features. First, we extract the TFIDF feature from the title, ingredients and cooking instructions of recipes. By determining the significance of word sequences through combining LSTM learned features with their TFIDF features, we encode a recipe into a TFIDF weighted vector for capturing significant key terms and how such key terms are used in the corresponding cooking instructions. Second, we combine the recipe TFIDF feature with the recipe sequence feature extracted through two-stage LSTM networks, which is effective in capturing the unique relationship between a recipe and its associated image(s). Third, we further incorporate TFIDF enhanced category semantics to improve the mapping of image modality and to regulate the similarity loss function during the iterative learning of cross-modal joint embedding. Experiments on the benchmark dataset Recipe1M show the proposed approach outperforms the state-of-the-art approaches.
- Open Access
Promoting High Diversity Ensemble Learning with EnsembleBench
Ensemble learning is gaining renewed interests in recent years. This paper presents EnsembleBench, a holistic framework for evaluating and recommending high diversity and high accuracy ensembles. The design of EnsembleBench offers three novel features: (1) EnsembleBench introduces a set of quantitative metrics for assessing the quality of ensembles and for comparing alternative ensembles constructed for the same learning tasks. (2) EnsembleBench implements a suite of baseline diversity metrics and optimized diversity metrics for identifying and selecting ensembles with high diversity and high quality, making it an effective framework for benchmarking, evaluating and recommending high diversity model ensembles. (3) Four representative ensemble consensus methods are provided in the first release of EnsembleBench, enabling empirical study on the impact of consensus methods on ensemble accuracy. A comprehensive experimental evaluation on popular benchmark datasets demonstrates the utility and effectiveness of EnsembleBench for promoting high diversity ensembles and boosting the overall performance of selected ensembles.
Open AccessPromoting High Diversity Ensemble Learning with EnsembleBench
Ensemble learning is gaining renewed interests in recent years. This paper presents EnsembleBench, a holistic framework for evaluating and recommending high diversity and high accuracy ensembles. The design of EnsembleBench offers three novel features: (1) EnsembleBench introduces a set of quantitative metrics for assessing the quality of ensembles and for comparing alternative ensembles constructed for the same learning tasks. (2) EnsembleBench implements a suite of baseline diversity metrics and optimized diversity metrics for identifying and selecting ensembles with high diversity and high quality, making it an effective framework for benchmarking, evaluating and recommending high diversity model ensembles. (3) Four representative ensemble consensus methods are provided in the first release of EnsembleBench, enabling empirical study on the impact of consensus methods on ensemble accuracy. A comprehensive experimental evaluation on popular benchmark datasets demonstrates the utility and effectiveness of EnsembleBench for promoting high diversity ensembles and boosting the overall performance of selected ensembles.
- 1 May 2020
- Published by Elsevier in Pattern Recognition Letters
- Vol. 133, 70-76
Image-to-video person re-identification with cross-modal embeddings
- 1 May 2020
- Published by Elsevier in Pattern Recognition Letters
- Vol. 133, 70-76
Image-to-video person re-identification with cross-modal embeddings
- 6 September 2018
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 78 (21) , 30769-30791
Enhancing multimodal deep representation learning by fixed model reuse
- 6 September 2018
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 78 (21) , 30769-30791
Enhancing multimodal deep representation learning by fixed model reuse
- Open Access
- 22 June 2015
- Published by Association for Computing Machinery (ACM)
- p. 1-12
WeChecker
Due to the rapid increase of Android apps and their wide usage to handle personal data, a precise and large-scaling checker is in need to validate the apps' permission flow before they are listed on the market. Several tools have been proposed to detect sensitive data leaks in Android apps. But these tools are not applicable to large-scale analysis since they fail to deal with the arbitrary execution orders of different event handlers smartly. Event handlers are invoked by the framework based on the system state, therefore we cannot pre-determine their order of execution. Besides, since all exported components can be invoked by an external app, the execution orders of these components are also arbitrary. A naive way to simulate these two types of arbitrary execution orders yields a permutation of all event handlers in an app. The time complexity is O(n!) where n is the number of event handlers in an app. This leads to a high analysis overhead when n is big. To give an illustration, CHEX [10] found 50.73 entry points of 44 unique class types in an app on average. In this paper we propose an improved static taint analysis to deal with the challenge brought by the arbitrary execution orders without sacrificing the high precision. Our analysis does not need to make permutations and achieves a polynomial time complexity. We also propose to unify the array and map access with object reference by propagating access paths to reduce the number of false positives due to field-insensitivity and over approximation of array access and map access. We implement a tool, WeChecker, to detect privilege escalation vulnerabilities [7] in Android apps. WeChecker achieves 96% precision and 96% recall in the state-of-the-art test suite DriodBench (for compairson, the precision and recall of FlowDroid [1] are 86% and 93%, respectively). The evaluation of WeChecker on real apps shows that it is efficient (average analysis time of each app: 29.985s) and fits for large-scale checking.
- 22 June 2015
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessWeChecker
Due to the rapid increase of Android apps and their wide usage to handle personal data, a precise and large-scaling checker is in need to validate the apps' permission flow before they are listed on the market. Several tools have been proposed to detect sensitive data leaks in Android apps. But these tools are not applicable to large-scale analysis since they fail to deal with the arbitrary execution orders of different event handlers smartly. Event handlers are invoked by the framework based on the system state, therefore we cannot pre-determine their order of execution. Besides, since all exported components can be invoked by an external app, the execution orders of these components are also arbitrary. A naive way to simulate these two types of arbitrary execution orders yields a permutation of all event handlers in an app. The time complexity is O(n!) where n is the number of event handlers in an app. This leads to a high analysis overhead when n is big. To give an illustration, CHEX [10] found 50.73 entry points of 44 unique class types in an app on average. In this paper we propose an improved static taint analysis to deal with the challenge brought by the arbitrary execution orders without sacrificing the high precision. Our analysis does not need to make permutations and achieves a polynomial time complexity. We also propose to unify the array and map access with object reference by propagating access paths to reduce the number of false positives due to field-insensitivity and over approximation of array access and map access. We implement a tool, WeChecker, to detect privilege escalation vulnerabilities [7] in Android apps. WeChecker achieves 96% precision and 96% recall in the state-of-the-art test suite DriodBench (for compairson, the precision and recall of FlowDroid [1] are 86% and 93%, respectively). The evaluation of WeChecker on real apps shows that it is efficient (average analysis time of each app: 29.985s) and fits for large-scale checking.