TÓM TẮT: Rút gọn thuộc tính là bài toán quan trọng trong bước tiền xử lý dữ liệu của quá trình kh... more TÓM TẮT: Rút gọn thuộc tính là bài toán quan trọng trong bước tiền xử lý dữ liệu của quá trình khai phá dữ liệu và khám phá tri thức. Trong mấy năm gần đây, các nhà nghiên cứu đề xuất các phương pháp rút gọn thuộc tính trực tiếp trên bảng quyết định gốc theo tiếp cận tập thô mờ (Fuzzy Rough Set FRS) nhằm nâng cao độ chính xác mô hình phân lớp. Tuy nhiên, số lượng thuộc tính thu được theo tiếp cận FRS chưa tối ưu do ràng buộc giữa các đối tượng trong bảng quyết định chưa được xem xét đầy đủ. Trong bài báo này, chúng tôi đề xuất phương pháp rút gọn thuộc tính trực tiếp trên bảng quyết định gốc theo tiếp cận tập thô mờ trực cảm (Intuitionistic Fuzzy Rough Set IFRS) dựa trên các đề xuất mới về hàm thành viên và không thành viên. Kết quả thử nghiệm trên các bộ dữ liệu mẫu cho thấy, số lượng thuộc tính của tập rút gọn theo phương pháp đề xuất giảm đáng kể so với các phương pháp FRS và một số phương pháp IFRS khác.
FAIR - NGHIÊN CỨU CƠ BẢN VÀ ỨNG DỤNG CÔNG NGHỆ THÔNG TIN - 2016, Aug 25, 2017
Traditional rough set based attribute reduction methods has performed on the decision tables with... more Traditional rough set based attribute reduction methods has performed on the decision tables with discretized value attribute domain. In recent years, many researchers has proposed some attribute reduction methods on the decision table with real attribute value domain based on fuzzy rough set. In this paper, we propose an attribute reduction method which performs directly on the decision table with real value domain using fuzzy distance. The experiment from UCI data sets showed that the accuracy classification of the proposed method is more efficient than the ones based on fuzzy positive region and fuzzy entropy.
In rough set theory, the number of all reducts for a given decision table can be exponential with... more In rough set theory, the number of all reducts for a given decision table can be exponential with respect to the number of attributes. This paper investigates the problem of determining the set of all reductive attributes which are present in at least one reduct of an incomplete decision table. We theoretically prove that this problem can be solved in polynomial time. This result shows that the problem of determining the union of all reducts can be solved in polynomial time, and the problem of determining the set of all redundant attributes which are not present in any reducts can also be solved in polynomial time.
Journal of Research and Development on Information and Communication Technology, 2016
Feature selection is a crucial problem need to be effectively solved in knowledge discovery ... more Feature selection is a crucial problem need to be effectively solved in knowledge discovery in databases because of two basic reasons: to minimize cost and to accurately classify data. Feature selection using rough set theory also called attribute reduction have attracted much attention from researchers and many results are gained. However, attribute reduction in dynamic databases is still in the first stage. This paper focus on develop incremental methods and algorithms to derive reducts hiring a distance measure when adding, deleting or updating objects. Since not re-implement the algorithms on the varied universal set, our algorithms significantly reduce the complexity of implementation time.
Tolerance rough set model is an effective tool for attribute reduction in incomplete decision tab... more Tolerance rough set model is an effective tool for attribute reduction in incomplete decision tables. In recent years, some incremental algorithms have been proposed to find reduct of dynamic incomplete decision tables in order to reduce computation time. However, they are classical filter algorithms, in which the classification accuracy of decision tables is computed after obtaining reduct. Therefore, the obtained reducts of these algorithms are not optimal on cardinality of reduct and classification accuracy. In this paper, we propose the incremental filter-wrapper algorithm IDS_IFW_AO to find one reduct of an incomplete desision table in case of adding multiple objects. The experimental results on some sample datasets show that the proposed filter-wrapper algorithm IDS_IFW_AO is more effective than the filter algorithm IARM-I [17] on classification accuracy and cardinality of reduct
In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information gr... more In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information granules and fuzzy information granularity is used to measure the granulation degree of a fuzzy granular structure. In general, the fuzzy information granularity characterizes discernibility ability among fuzzy information granules in a fuzzy granular structure. In recent years, researchers have proposed some concepts of fuzzy information granularity based on partial order relations. However, the existing forms of fuzzy information granularity have some limitations when evaluating the fineness/coarseness between two fuzzy granular structures. In this paper, we propose an extension of fuzzy information granularity based on a fuzzy distance measure. We prove theoretically and experimentally that the proposed fuzzy information granularity is the best one to distinguish fuzzy granular structures.
In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information gr... more In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information granules and fuzzy information granularity is used to measure the granulation degree of a fuzzy granular structure. In general, the fuzzy information granularity characterizes discernibility ability among fuzzy information granules in a fuzzy granular structure. In recent years, researchers have proposed some concepts of fuzzy information granularity based on partial order relations. However, the existing forms of fuzzy information granularity have some limitations when evaluating the fineness/coarseness between two fuzzy granular structures. In this paper, we propose an extension of fuzzy information granularity based on a fuzzy distance measure. We prove theoretically and experimentally that the proposed fuzzy information granularity is the best one to distinguish fuzzy granular structures. ACM Computing Classification System (1998): I.5.2, I.2.6.
family proteins, the cytotoxic effects of these BH3 peptides can be reduced in certain cancer cel... more family proteins, the cytotoxic effects of these BH3 peptides can be reduced in certain cancer cells. We recently found that the amphipathic tail-anchoring peptide (ATAP) from Bfl-1, a bifunctional Bcl-2 family member, displayed strong pro-apoptotic activity by permeabilizing the mitochondrial outer membrane (MOM). Here we tested if the activity of ATAP requires other cellular factors and whether ATAP has an advantage over the BH3 peptides in targeting cancer cells. We reconstituted the membrane permeabilizing activity of ATAP in liposomes and found that ATAP rapidly released fluorescent molecules of the size of cytochrome c, suggesting that ATAP membrane permeabilizing activity is independent of other protein factors. Confocal microscopic imaging revealed specific targeting of ATAP to MOM, whereas BH3-peptides showed diffuse cytosolic distribution. While the pro-apoptotic activity of BH3 peptides was largely inhibited by either overexpression of Bcl-2 or Bcl-x L or nullification of Bax and Bak in cells, the apoptotic function of ATAP was not affected by these cellular factors. Since ATAP can target to mitochondria membrane and its potent apoptotic activity does not dependent on the content of Bcl-2 family proteins, it represents a promising lead for a new class of anti-cancer drugs that can potentially overcome the intrinsic apoptosis-resistant nature of cancer cells.
TÓM TẮT: Rút gọn thuộc tính là bài toán quan trọng trong bước tiền xử lý dữ liệu của quá trình kh... more TÓM TẮT: Rút gọn thuộc tính là bài toán quan trọng trong bước tiền xử lý dữ liệu của quá trình khai phá dữ liệu và khám phá tri thức. Trong mấy năm gần đây, các nhà nghiên cứu đề xuất các phương pháp rút gọn thuộc tính trực tiếp trên bảng quyết định gốc theo tiếp cận tập thô mờ (Fuzzy Rough Set FRS) nhằm nâng cao độ chính xác mô hình phân lớp. Tuy nhiên, số lượng thuộc tính thu được theo tiếp cận FRS chưa tối ưu do ràng buộc giữa các đối tượng trong bảng quyết định chưa được xem xét đầy đủ. Trong bài báo này, chúng tôi đề xuất phương pháp rút gọn thuộc tính trực tiếp trên bảng quyết định gốc theo tiếp cận tập thô mờ trực cảm (Intuitionistic Fuzzy Rough Set IFRS) dựa trên các đề xuất mới về hàm thành viên và không thành viên. Kết quả thử nghiệm trên các bộ dữ liệu mẫu cho thấy, số lượng thuộc tính của tập rút gọn theo phương pháp đề xuất giảm đáng kể so với các phương pháp FRS và một số phương pháp IFRS khác.
FAIR - NGHIÊN CỨU CƠ BẢN VÀ ỨNG DỤNG CÔNG NGHỆ THÔNG TIN - 2016, Aug 25, 2017
Traditional rough set based attribute reduction methods has performed on the decision tables with... more Traditional rough set based attribute reduction methods has performed on the decision tables with discretized value attribute domain. In recent years, many researchers has proposed some attribute reduction methods on the decision table with real attribute value domain based on fuzzy rough set. In this paper, we propose an attribute reduction method which performs directly on the decision table with real value domain using fuzzy distance. The experiment from UCI data sets showed that the accuracy classification of the proposed method is more efficient than the ones based on fuzzy positive region and fuzzy entropy.
In rough set theory, the number of all reducts for a given decision table can be exponential with... more In rough set theory, the number of all reducts for a given decision table can be exponential with respect to the number of attributes. This paper investigates the problem of determining the set of all reductive attributes which are present in at least one reduct of an incomplete decision table. We theoretically prove that this problem can be solved in polynomial time. This result shows that the problem of determining the union of all reducts can be solved in polynomial time, and the problem of determining the set of all redundant attributes which are not present in any reducts can also be solved in polynomial time.
Journal of Research and Development on Information and Communication Technology, 2016
Feature selection is a crucial problem need to be effectively solved in knowledge discovery ... more Feature selection is a crucial problem need to be effectively solved in knowledge discovery in databases because of two basic reasons: to minimize cost and to accurately classify data. Feature selection using rough set theory also called attribute reduction have attracted much attention from researchers and many results are gained. However, attribute reduction in dynamic databases is still in the first stage. This paper focus on develop incremental methods and algorithms to derive reducts hiring a distance measure when adding, deleting or updating objects. Since not re-implement the algorithms on the varied universal set, our algorithms significantly reduce the complexity of implementation time.
Tolerance rough set model is an effective tool for attribute reduction in incomplete decision tab... more Tolerance rough set model is an effective tool for attribute reduction in incomplete decision tables. In recent years, some incremental algorithms have been proposed to find reduct of dynamic incomplete decision tables in order to reduce computation time. However, they are classical filter algorithms, in which the classification accuracy of decision tables is computed after obtaining reduct. Therefore, the obtained reducts of these algorithms are not optimal on cardinality of reduct and classification accuracy. In this paper, we propose the incremental filter-wrapper algorithm IDS_IFW_AO to find one reduct of an incomplete desision table in case of adding multiple objects. The experimental results on some sample datasets show that the proposed filter-wrapper algorithm IDS_IFW_AO is more effective than the filter algorithm IARM-I [17] on classification accuracy and cardinality of reduct
In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information gr... more In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information granules and fuzzy information granularity is used to measure the granulation degree of a fuzzy granular structure. In general, the fuzzy information granularity characterizes discernibility ability among fuzzy information granules in a fuzzy granular structure. In recent years, researchers have proposed some concepts of fuzzy information granularity based on partial order relations. However, the existing forms of fuzzy information granularity have some limitations when evaluating the fineness/coarseness between two fuzzy granular structures. In this paper, we propose an extension of fuzzy information granularity based on a fuzzy distance measure. We prove theoretically and experimentally that the proposed fuzzy information granularity is the best one to distinguish fuzzy granular structures.
In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information gr... more In fuzzy granular computing, a fuzzy granular structure is the collection of fuzzy information granules and fuzzy information granularity is used to measure the granulation degree of a fuzzy granular structure. In general, the fuzzy information granularity characterizes discernibility ability among fuzzy information granules in a fuzzy granular structure. In recent years, researchers have proposed some concepts of fuzzy information granularity based on partial order relations. However, the existing forms of fuzzy information granularity have some limitations when evaluating the fineness/coarseness between two fuzzy granular structures. In this paper, we propose an extension of fuzzy information granularity based on a fuzzy distance measure. We prove theoretically and experimentally that the proposed fuzzy information granularity is the best one to distinguish fuzzy granular structures. ACM Computing Classification System (1998): I.5.2, I.2.6.
family proteins, the cytotoxic effects of these BH3 peptides can be reduced in certain cancer cel... more family proteins, the cytotoxic effects of these BH3 peptides can be reduced in certain cancer cells. We recently found that the amphipathic tail-anchoring peptide (ATAP) from Bfl-1, a bifunctional Bcl-2 family member, displayed strong pro-apoptotic activity by permeabilizing the mitochondrial outer membrane (MOM). Here we tested if the activity of ATAP requires other cellular factors and whether ATAP has an advantage over the BH3 peptides in targeting cancer cells. We reconstituted the membrane permeabilizing activity of ATAP in liposomes and found that ATAP rapidly released fluorescent molecules of the size of cytochrome c, suggesting that ATAP membrane permeabilizing activity is independent of other protein factors. Confocal microscopic imaging revealed specific targeting of ATAP to MOM, whereas BH3-peptides showed diffuse cytosolic distribution. While the pro-apoptotic activity of BH3 peptides was largely inhibited by either overexpression of Bcl-2 or Bcl-x L or nullification of Bax and Bak in cells, the apoptotic function of ATAP was not affected by these cellular factors. Since ATAP can target to mitochondria membrane and its potent apoptotic activity does not dependent on the content of Bcl-2 family proteins, it represents a promising lead for a new class of anti-cancer drugs that can potentially overcome the intrinsic apoptosis-resistant nature of cancer cells.
Uploads
Papers by Nguyen Giang