Li Weigang | Universidade de Brasília - UnB (original) (raw)

Papers by Li Weigang

IEEE Latin America Transactions

Advances in Intelligent Systems and Computing, 2019

Implementation of Trajectory Based Operations (TBO) has been updating the structure of the advanc... more Implementation of Trajectory Based Operations (TBO) has been updating the structure of the advanced Air Traffic Management (ATM). Although several methodologies for conflict detection and resolution (CDR) have been developed to the aviation community, the legacy problem is to find an efficient scheme to present the trajectories in this complex network with massive data and further to detect and resolve the conflicts. In this research we develop a CDR framework based on the management of predicted 4D-trajectories using a Not Only SQL (NoSQL) database and local search algorithms for conflict resolution. This paper describes the architecture and algorithms of the proposed solution in 4-Dimensional Trajectory (4DT). With the application of Trajectory Prediction (TP) simulator using the Brazilian flight plan database, the results from case study show the effectiveness of the proposed methods for this sophisticated problem in ATM.

WIT Transactions on Information and Communication Technologies, 1970

A Know ledge-Based System (KBS) has been designed for the Air Traffic Flow Management (ATFM) prob... more A Know ledge-Based System (KBS) has been designed for the Air Traffic Flow Management (ATFM) problem. For timetable rescheduling, the system tends to modify airlines timetable to smooth traffic peaks at airports during rush-hours. For centralized flow control, the system works on-line to forecast the place, time and magnitude of the congestions and to take some actions to prevent these congestions. As an Artificial Intelligence language, Prolog was chosen to develop the prototype of the Knowledge-Based ATFM System. By using this system, Brazilian ATFM, which includes the principal airports of this country, has been studied.

Security, Privacy, and Anonymity in Computation, Communication, and Storage, 2017

This paper briefly reviews the state-of-the-art in Artificial Intelligence (AI) applied to Air Tr... more This paper briefly reviews the state-of-the-art in Artificial Intelligence (AI) applied to Air Traffic Management (ATM). The research topics include the application of semantic ontology, multi-agent systems, reinforcement learning (RL), and game theory in ATM. Likewise, this paper also highlights our research advances in this area. In this case, we describe a new Probabilistic Web Ontology Language (PR-OWL) algorithm to enable the reasoning on big datasets in polynomial time. Then, we present the use of both Particle Swarm Optimization (PSO) and Simulated Annealing (SA) algorithms in 4D trajectory management. Next, we describe the usage of Multi-agent Planning (MAP) theory on airport ground handling management. Finally, this paper envisions some research and development directions of AI applied to ATM. It includes: (a) mapping and reducing the gaps between advanced AI technologies and ATM; (b) considering uncertainty in Semantic Ontology for SWIM data exchanging models in ATM; (c) using big data analytics in SWIM; and (d) integrating collaborative ATM technologies towards intelligent SWIM (I-SWIM).

Inaccurate carbon emissions predictions may be one of the root factors leading to the overall ine... more Inaccurate carbon emissions predictions may be one of the root factors leading to the overall ineffectiveness of the European Union environmental regulatory framework. Therefore, the present article aims at introducing a novel computational learning framework for carbon emissions prediction incorporating a RReliefF driven features selection and an iterative neural network architecture improvement. Our learning framework algorithmic architecture iteratively chains the features selection process and the backpropagation artificial neural network architecture design based on the data assessment accomplished by the RReliefF algorithm. Thus a better features set / neural network architecture combination is obtained for each specific prediction target. The implemented framework was trained and tested with real world data obtained from the European Union, International Energy Agency, Organisation for Economic Co-operation and Development, and World Bank, for the period 1990–2017. The framew...

Classification of text documents is of significant importance in the field of data mining and mac... more Classification of text documents is of significant importance in the field of data mining and machine learning. However, the vector representation of documents, in classification problems, results in a highly sparse data with immense number of variables. This necessitates applying an efficient variables selection and dimensionality reduction technique that ensures model’s selectivity, accuracy and robustness with fewer variables. This paper introduces a new coefficient, the Variables Strength Coefficient (VSC), which permits retaining variables with strong Modeling and Discriminatory powers. A variable with VSC greater than a predefined threshold is considered to have strong power in both modeling data and discriminating classes and thus retained, while weaker variables are discarded. This straightforward technique results in maximizing the differences between classes while preserving the modeling power of variables. This paper also proposes applying a classification technique that ...

ISPRS International Journal of Geo-Information, 2021

Misappropriation of public lands is an ongoing government concern. In Brazil, the beach zone is p... more Misappropriation of public lands is an ongoing government concern. In Brazil, the beach zone is public property, but many private establishments use it for economic purposes, requiring constant inspection. Among the undue targets, the individual mapping of straw beach umbrellas (SBUs) attached to the sand is a great challenge due to their small size, high presence, and agglutinated appearance. This study aims to automatically detect and count SBUs on public beaches using high-resolution images and instance segmentation, obtaining pixel-wise semantic information and individual object detection. This study is the first instance segmentation application on coastal areas and the first using WorldView-3 (WV-3) images. We used the Mask-RCNN with some modifications: (a) multispectral input for the WorldView3 imagery (eight channels), (b) improved the sliding window algorithm for large image classification, and (c) comparison of different image resizing ratios to improve small object detect...

2015 18th International Conference on Information Fusion (Fusion), 2015

Twitter is an online social networking (OSN) service that enables users to send and read short me... more Twitter is an online social networking (OSN) service that enables users to send and read short messages called “tweets”. As of December 2014, Twitter has more than 500 million users, out of which more than 284 million are active users and about 500 million tweets are posted every day. Tweet sentiment analysis (TSA) identifies a valuable platform for the OSN study which provides insights into the opinion of the public about culture, products and political agendas and thereby is able to predict the trends in specific domains. In order to execute efficient TSA on a particular topic or domain, a TSA approach with unified tool, UnB TSA, is proposed consisting of four steps: tweets collection, refinement (excluding noisy tweets), sentiment lexicon creation and sentiment analysis. As a key part, the lexicon is domain-specific that incorporates expressions whose sentiment varies from one domain to another. Four algorithms including expanding limited hashtags into a larger and more complete ...

Lecture Notes in Business Information Processing, 2015

Advances in intelligent systems and computing, Jul 5, 2018

The Collaborative Trajectory Options Program (CTOP) makes each airline possible to share its rout... more The Collaborative Trajectory Options Program (CTOP) makes each airline possible to share its route options to air traffic control center, and so achieve better business goals by reducing strategic operational costs. In Brazil, there are initial efforts to verify the benefits of CTOP implementation to improve the air traffic fluency and financial results. This paper presents a novel approach for Brazilian airspace using Genetic Algorithms to decrease the delay between available slots during CTOP. The slot optimization keeps improving in a safety-separating window of each aircraft en route. The case study presented an reducement about 70% of delay of a certain airline, when used this decision support system by air traffic control authority.

2022 4th International Conference on Natural Language Processing (ICNLP), Mar 1, 2022

Communications in computer and information science, 2018

Air traffic management (ATM) has increased the complexity of computational solutions over the las... more Air traffic management (ATM) has increased the complexity of computational solutions over the last decades to support the discovery process and knowledge management in decision-making process. It involves dealing with more information about flights, weather, delays, forecasting scenarios, and others. An evolution of ATM programs was the Collaborative Trajectory Options Program (CTOP), which is applied in the USA since 2014. In Brazil, some initial studies are analyzing this evolution and how the current air traffic management could be improved. One option is to implement the concept addressed by CTOP and its specificities based on Brazil context. This paper presents a computational solution that uses the main concept of CTOP with a collaborative approach to implement the program in Brazil. The main objective is to reduce delays of flights captured by the CTOP, so that the airlines collaboratively reduce delays in their flights and each airline can their business goals, using techniques of Artificial Intelligence with Game Theory and Folk Theorem. The CTOP approach to Brazilian air traffic management is a first initiative to improve the air traffic programs by using the collaborative decision-making process to achieve better results. The achieved results were important for this initial case study, decreasing about 18% of delays in CTOP captured flights, when it was considered only the priority flights of an airline.

Applied Sciences

The most common method used to document monetary transactions in Brazil is by issuing electronic ... more The most common method used to document monetary transactions in Brazil is by issuing electronic invoices (NF-e). The audit of electronic invoices is essential, and this can be improved by using data mining solutions, such as clustering and anomaly detection. However, applying these solutions is not a simple task because NF-e data contains millions of records with noisy fields and nonstandard documents, especially short text descriptions. In addition to these challenges, it is costly to extract information from short texts to identify traces of mismanagement, embezzlement, commercial fraud or tax evasion. Analyzing such data can be more effective when divided into well-defined groups. However, efficient solutions for clustering data with characteristics similar to NF-es have not yet been proposed in the literature. We developed ELINAC, a service for clustering short-text data in NF-es that uses an automatic encoder to cluster data. ELINAC aids in auditing transactions documented in ...

2015 18th International Conference on Information Fusion (Fusion), 2015

The probabilistic ontology language PR-OWL (Probabilistic OWL) uses Multi-Entity Bayesian Network... more The probabilistic ontology language PR-OWL (Probabilistic OWL) uses Multi-Entity Bayesian Networks (MEBN), an extension of Bayesian networks with first-order logic, to add the ability to deal with uncertainty to OWL, the main language of the Semantic Web. A second version, PR-OWL 2, was proposed to allow the construction of hybrid ontologies, containing deterministic and probabilistic parts. Existing PROWL implementations cannot deal with very large assertive databases. This limitation is a main obstacle for applying the language in real domains, such as Maritime Domain Awareness (MDA). This paper proposes a PR-OWL extension using RDF triplestores and the OWL 2 RL profile, based on rules, in order to allow dealing with uncertainty in ontologies with millions of assertions. We illustrate our ideas with an MDA ontology built for the PROGNOS (PRobabilistic OntoloGies for Net-centric Operation Systems) project.

Aircraft fly in an environment that is subject to constant weather changes, which considerably in... more Aircraft fly in an environment that is subject to constant weather changes, which considerably influences the decision-making process in Air Traffic Management (ATM). The stakeholders in ATM track weather conditions to appropriately respond against new environmental settings. The proposed work builds an intelligent system to constantly monitor the impact of severe weather on airways, which are corridors with specific width and height connecting two locations in the airspace. The proposed approach integrates weather information on convection cells obtained from ground-based weather radars, and flight tracking information detailing flight positions in real time. To delimit the boundaries of airways, the set of flight positions is transformed to a more convenient one using linear interpolation. Then, a cluster analysis via Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is performed into this new set. The algorithm compares the positions of the clusters found with ...

Multi-Entity Bayesian Network (MEBN) is an expressive first-order probabilistic logic that repres... more Multi-Entity Bayesian Network (MEBN) is an expressive first-order probabilistic logic that represents the domain using parameterized fragments of Bayesian networks. Probabilistic-OWL (PR-OWL) uses MEBN to add uncertainty support to OWL, the main language of the Semantic Web. The reasoning in MEBN is made by the construction of a Situation-Specific Bayesian Network (SSBN), a minimal Bayesian network sufficient to compute the response to queries. A Bottom-Up algorithm has been proposed for generating SSBNs in MEBN. However, this approach presents scalability problems since the algorithm starts from all the query and evidence nodes, which can be a very large set in real domains. To address this problem, we present a new scalable algorithm for generating SSBNs based on the Bayes-Ball method, a well-known and efficient algorithm for discovering d-separated nodes of target sets in Bayesian networks. The novel SSBN algorithm used together with Resource Description Framework (RDF) databases...

IEEE Latin America Transactions

Advances in Intelligent Systems and Computing, 2019

Implementation of Trajectory Based Operations (TBO) has been updating the structure of the advanc... more Implementation of Trajectory Based Operations (TBO) has been updating the structure of the advanced Air Traffic Management (ATM). Although several methodologies for conflict detection and resolution (CDR) have been developed to the aviation community, the legacy problem is to find an efficient scheme to present the trajectories in this complex network with massive data and further to detect and resolve the conflicts. In this research we develop a CDR framework based on the management of predicted 4D-trajectories using a Not Only SQL (NoSQL) database and local search algorithms for conflict resolution. This paper describes the architecture and algorithms of the proposed solution in 4-Dimensional Trajectory (4DT). With the application of Trajectory Prediction (TP) simulator using the Brazilian flight plan database, the results from case study show the effectiveness of the proposed methods for this sophisticated problem in ATM.

WIT Transactions on Information and Communication Technologies, 1970

A Know ledge-Based System (KBS) has been designed for the Air Traffic Flow Management (ATFM) prob... more A Know ledge-Based System (KBS) has been designed for the Air Traffic Flow Management (ATFM) problem. For timetable rescheduling, the system tends to modify airlines timetable to smooth traffic peaks at airports during rush-hours. For centralized flow control, the system works on-line to forecast the place, time and magnitude of the congestions and to take some actions to prevent these congestions. As an Artificial Intelligence language, Prolog was chosen to develop the prototype of the Knowledge-Based ATFM System. By using this system, Brazilian ATFM, which includes the principal airports of this country, has been studied.

Security, Privacy, and Anonymity in Computation, Communication, and Storage, 2017

This paper briefly reviews the state-of-the-art in Artificial Intelligence (AI) applied to Air Tr... more This paper briefly reviews the state-of-the-art in Artificial Intelligence (AI) applied to Air Traffic Management (ATM). The research topics include the application of semantic ontology, multi-agent systems, reinforcement learning (RL), and game theory in ATM. Likewise, this paper also highlights our research advances in this area. In this case, we describe a new Probabilistic Web Ontology Language (PR-OWL) algorithm to enable the reasoning on big datasets in polynomial time. Then, we present the use of both Particle Swarm Optimization (PSO) and Simulated Annealing (SA) algorithms in 4D trajectory management. Next, we describe the usage of Multi-agent Planning (MAP) theory on airport ground handling management. Finally, this paper envisions some research and development directions of AI applied to ATM. It includes: (a) mapping and reducing the gaps between advanced AI technologies and ATM; (b) considering uncertainty in Semantic Ontology for SWIM data exchanging models in ATM; (c) using big data analytics in SWIM; and (d) integrating collaborative ATM technologies towards intelligent SWIM (I-SWIM).

Inaccurate carbon emissions predictions may be one of the root factors leading to the overall ine... more Inaccurate carbon emissions predictions may be one of the root factors leading to the overall ineffectiveness of the European Union environmental regulatory framework. Therefore, the present article aims at introducing a novel computational learning framework for carbon emissions prediction incorporating a RReliefF driven features selection and an iterative neural network architecture improvement. Our learning framework algorithmic architecture iteratively chains the features selection process and the backpropagation artificial neural network architecture design based on the data assessment accomplished by the RReliefF algorithm. Thus a better features set / neural network architecture combination is obtained for each specific prediction target. The implemented framework was trained and tested with real world data obtained from the European Union, International Energy Agency, Organisation for Economic Co-operation and Development, and World Bank, for the period 1990–2017. The framew...

Classification of text documents is of significant importance in the field of data mining and mac... more Classification of text documents is of significant importance in the field of data mining and machine learning. However, the vector representation of documents, in classification problems, results in a highly sparse data with immense number of variables. This necessitates applying an efficient variables selection and dimensionality reduction technique that ensures model’s selectivity, accuracy and robustness with fewer variables. This paper introduces a new coefficient, the Variables Strength Coefficient (VSC), which permits retaining variables with strong Modeling and Discriminatory powers. A variable with VSC greater than a predefined threshold is considered to have strong power in both modeling data and discriminating classes and thus retained, while weaker variables are discarded. This straightforward technique results in maximizing the differences between classes while preserving the modeling power of variables. This paper also proposes applying a classification technique that ...

ISPRS International Journal of Geo-Information, 2021

Misappropriation of public lands is an ongoing government concern. In Brazil, the beach zone is p... more Misappropriation of public lands is an ongoing government concern. In Brazil, the beach zone is public property, but many private establishments use it for economic purposes, requiring constant inspection. Among the undue targets, the individual mapping of straw beach umbrellas (SBUs) attached to the sand is a great challenge due to their small size, high presence, and agglutinated appearance. This study aims to automatically detect and count SBUs on public beaches using high-resolution images and instance segmentation, obtaining pixel-wise semantic information and individual object detection. This study is the first instance segmentation application on coastal areas and the first using WorldView-3 (WV-3) images. We used the Mask-RCNN with some modifications: (a) multispectral input for the WorldView3 imagery (eight channels), (b) improved the sliding window algorithm for large image classification, and (c) comparison of different image resizing ratios to improve small object detect...

2015 18th International Conference on Information Fusion (Fusion), 2015

Twitter is an online social networking (OSN) service that enables users to send and read short me... more Twitter is an online social networking (OSN) service that enables users to send and read short messages called “tweets”. As of December 2014, Twitter has more than 500 million users, out of which more than 284 million are active users and about 500 million tweets are posted every day. Tweet sentiment analysis (TSA) identifies a valuable platform for the OSN study which provides insights into the opinion of the public about culture, products and political agendas and thereby is able to predict the trends in specific domains. In order to execute efficient TSA on a particular topic or domain, a TSA approach with unified tool, UnB TSA, is proposed consisting of four steps: tweets collection, refinement (excluding noisy tweets), sentiment lexicon creation and sentiment analysis. As a key part, the lexicon is domain-specific that incorporates expressions whose sentiment varies from one domain to another. Four algorithms including expanding limited hashtags into a larger and more complete ...

Lecture Notes in Business Information Processing, 2015

Advances in intelligent systems and computing, Jul 5, 2018

The Collaborative Trajectory Options Program (CTOP) makes each airline possible to share its rout... more The Collaborative Trajectory Options Program (CTOP) makes each airline possible to share its route options to air traffic control center, and so achieve better business goals by reducing strategic operational costs. In Brazil, there are initial efforts to verify the benefits of CTOP implementation to improve the air traffic fluency and financial results. This paper presents a novel approach for Brazilian airspace using Genetic Algorithms to decrease the delay between available slots during CTOP. The slot optimization keeps improving in a safety-separating window of each aircraft en route. The case study presented an reducement about 70% of delay of a certain airline, when used this decision support system by air traffic control authority.

2022 4th International Conference on Natural Language Processing (ICNLP), Mar 1, 2022

Communications in computer and information science, 2018

Air traffic management (ATM) has increased the complexity of computational solutions over the las... more Air traffic management (ATM) has increased the complexity of computational solutions over the last decades to support the discovery process and knowledge management in decision-making process. It involves dealing with more information about flights, weather, delays, forecasting scenarios, and others. An evolution of ATM programs was the Collaborative Trajectory Options Program (CTOP), which is applied in the USA since 2014. In Brazil, some initial studies are analyzing this evolution and how the current air traffic management could be improved. One option is to implement the concept addressed by CTOP and its specificities based on Brazil context. This paper presents a computational solution that uses the main concept of CTOP with a collaborative approach to implement the program in Brazil. The main objective is to reduce delays of flights captured by the CTOP, so that the airlines collaboratively reduce delays in their flights and each airline can their business goals, using techniques of Artificial Intelligence with Game Theory and Folk Theorem. The CTOP approach to Brazilian air traffic management is a first initiative to improve the air traffic programs by using the collaborative decision-making process to achieve better results. The achieved results were important for this initial case study, decreasing about 18% of delays in CTOP captured flights, when it was considered only the priority flights of an airline.

Applied Sciences

The most common method used to document monetary transactions in Brazil is by issuing electronic ... more The most common method used to document monetary transactions in Brazil is by issuing electronic invoices (NF-e). The audit of electronic invoices is essential, and this can be improved by using data mining solutions, such as clustering and anomaly detection. However, applying these solutions is not a simple task because NF-e data contains millions of records with noisy fields and nonstandard documents, especially short text descriptions. In addition to these challenges, it is costly to extract information from short texts to identify traces of mismanagement, embezzlement, commercial fraud or tax evasion. Analyzing such data can be more effective when divided into well-defined groups. However, efficient solutions for clustering data with characteristics similar to NF-es have not yet been proposed in the literature. We developed ELINAC, a service for clustering short-text data in NF-es that uses an automatic encoder to cluster data. ELINAC aids in auditing transactions documented in ...

2015 18th International Conference on Information Fusion (Fusion), 2015

The probabilistic ontology language PR-OWL (Probabilistic OWL) uses Multi-Entity Bayesian Network... more The probabilistic ontology language PR-OWL (Probabilistic OWL) uses Multi-Entity Bayesian Networks (MEBN), an extension of Bayesian networks with first-order logic, to add the ability to deal with uncertainty to OWL, the main language of the Semantic Web. A second version, PR-OWL 2, was proposed to allow the construction of hybrid ontologies, containing deterministic and probabilistic parts. Existing PROWL implementations cannot deal with very large assertive databases. This limitation is a main obstacle for applying the language in real domains, such as Maritime Domain Awareness (MDA). This paper proposes a PR-OWL extension using RDF triplestores and the OWL 2 RL profile, based on rules, in order to allow dealing with uncertainty in ontologies with millions of assertions. We illustrate our ideas with an MDA ontology built for the PROGNOS (PRobabilistic OntoloGies for Net-centric Operation Systems) project.

Aircraft fly in an environment that is subject to constant weather changes, which considerably in... more Aircraft fly in an environment that is subject to constant weather changes, which considerably influences the decision-making process in Air Traffic Management (ATM). The stakeholders in ATM track weather conditions to appropriately respond against new environmental settings. The proposed work builds an intelligent system to constantly monitor the impact of severe weather on airways, which are corridors with specific width and height connecting two locations in the airspace. The proposed approach integrates weather information on convection cells obtained from ground-based weather radars, and flight tracking information detailing flight positions in real time. To delimit the boundaries of airways, the set of flight positions is transformed to a more convenient one using linear interpolation. Then, a cluster analysis via Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is performed into this new set. The algorithm compares the positions of the clusters found with ...

Multi-Entity Bayesian Network (MEBN) is an expressive first-order probabilistic logic that repres... more Multi-Entity Bayesian Network (MEBN) is an expressive first-order probabilistic logic that represents the domain using parameterized fragments of Bayesian networks. Probabilistic-OWL (PR-OWL) uses MEBN to add uncertainty support to OWL, the main language of the Semantic Web. The reasoning in MEBN is made by the construction of a Situation-Specific Bayesian Network (SSBN), a minimal Bayesian network sufficient to compute the response to queries. A Bottom-Up algorithm has been proposed for generating SSBNs in MEBN. However, this approach presents scalability problems since the algorithm starts from all the query and evidence nodes, which can be a very large set in real domains. To address this problem, we present a new scalable algorithm for generating SSBNs based on the Bayes-Ball method, a well-known and efficient algorithm for discovering d-separated nodes of target sets in Bayesian networks. The novel SSBN algorithm used together with Resource Description Framework (RDF) databases...