Rank reversals appearing in AHP are discussed. It is proved with examples that they are in certain degree universal in the process of decision making. The mechanism of rank reversal is expounded. It is believed that...Rank reversals appearing in AHP are discussed. It is proved with examples that they are in certain degree universal in the process of decision making. The mechanism of rank reversal is expounded. It is believed that rank reversal can not deny the reasonableness of AHP and the axioms of independences of irrelevant alternatives.展开更多
This paper proposed a novel feature selection method LUIFS （ latent utility of irrelevant feature selection） that not only selects the relevant features, but also targets at discovering the latent useful irrelevant ...This paper proposed a novel feature selection method LUIFS （ latent utility of irrelevant feature selection） that not only selects the relevant features, but also targets at discovering the latent useful irrelevant attributes by measuring their supportive importance to other attributes. The method minimizes the information lost and simultaneously maximizes the final classification accuracy. The classification error rates of the LUIFS method on 16 real-life datasets from UCI machine learning repository were evaluated using the ID3, Na^ve-Bayes, and IB （instance-based classifier） learning algorithms, respectively; and compared with those of the same algorithms with no feature selection （NoFS）, feature subset selection （FSS）, and correlation-based feature selection （CFS）. The empirical results demonstrate that the LUIFS can improve the performance of learning algorithms by taking the latent relevance for irrelevant attributes into consideration, and hence including those potentially important attributes into the optimal feature subset for classification.展开更多
The conception of truth-making,albeit in a rudimentary form,could already be discerned in the writings of G.E.Moore and E.Husserl in the early 1900s.A few years later it was more extensively exploited by William James...The conception of truth-making,albeit in a rudimentary form,could already be discerned in the writings of G.E.Moore and E.Husserl in the early 1900s.A few years later it was more extensively exploited by William James.It was Wittgenstein,however,who gave the concept a precise meaning.In 1913/1914 Wittgenstein advanced a theory of possible worlds,only one of which was real.Every proposition suggests a part of a possible world which does or does not correspond to parts of the real world.In the first case the proposition is true,in the second case false.Moreover,the part of the real world makes the sentence true.This part is a truth-maker,and the sentence is a truth-bearer.Surprisingly enough,Wittgenstein’s concept of truth-making had its family resemblance with William James’s conception of truth.In 1915 Wittgenstein stopped using the concept of truth-making-it was also not mentioned in the Tractatus.Unfortunately,Russell did not notice this and in 1918 he adopted the concept of truth-making.In the 1930s,it was used by some second generation analytic philosophers(Schlick,Stebbing,and Wisdom).However,it became rather popular among analytic philosophers only in the 1980s.展开更多
Employing a 4-form ansatz of 11-dimensional supergravity over a non-dynamical AdS_(4)×S^(7)/Z_(k)background and setting the internal space as an S1 Hopf fibration on CP3,we obtain a consistent truncation.The(pseu...Employing a 4-form ansatz of 11-dimensional supergravity over a non-dynamical AdS_(4)×S^(7)/Z_(k)background and setting the internal space as an S1 Hopf fibration on CP3,we obtain a consistent truncation.The(pseudo)scalars,in the resulting scalar equations in Euclidean AdS_(4)space,may be considered to arise from(anti)M-branes wrapping around the internal directions in the(Wick-rotated)skew-whiffed M2-brane background(as the resulting theory is for anti-M2-branes),thus realizing the modes after swapping the three fundamental representations 8_(s),8_(c),and 8_(v) of SO(8).Taking the backreaction on the external and internal spaces,we obtain the massless and massive modes,corresponding to exactly marginal and marginally irrelevant deformations on the boundary CFT3,respectively.Subsequently,we obtain a closed solution for the bulk equation and compute its correction with respect to the background action.Next,considering the Higgs-like(breathing)mode m^(2)=18,having all supersymmetries as well as parity and scale-invariance broken,solving the associated bulk equation with mathematical methods,specifically the Adomian decomposition method,and analyzing the behavior near the boundary of the solutions,we realize the boundary duals in the SU(4)×U(1)-singlet sectors of the ABJM model.Then,introducing the new dual deformationΔ_(+)=3,6 operators made of bi-fundamental scalars,fermions,and U(1)gauge fields,we obtain the SO(4)-invariant solutions as small instantons on a three-sphere with the radius at infinity,which correspond to collapsing bulk bubbles leading to big-crunch singularities.展开更多
文摘Rank reversals appearing in AHP are discussed. It is proved with examples that they are in certain degree universal in the process of decision making. The mechanism of rank reversal is expounded. It is believed that rank reversal can not deny the reasonableness of AHP and the axioms of independences of irrelevant alternatives.
基金The Science and Technology Development Fund from Macao Government (No007/2006/A)
文摘This paper proposed a novel feature selection method LUIFS （ latent utility of irrelevant feature selection） that not only selects the relevant features, but also targets at discovering the latent useful irrelevant attributes by measuring their supportive importance to other attributes. The method minimizes the information lost and simultaneously maximizes the final classification accuracy. The classification error rates of the LUIFS method on 16 real-life datasets from UCI machine learning repository were evaluated using the ID3, Na^ve-Bayes, and IB （instance-based classifier） learning algorithms, respectively; and compared with those of the same algorithms with no feature selection （NoFS）, feature subset selection （FSS）, and correlation-based feature selection （CFS）. The empirical results demonstrate that the LUIFS can improve the performance of learning algorithms by taking the latent relevance for irrelevant attributes into consideration, and hence including those potentially important attributes into the optimal feature subset for classification.
文摘The conception of truth-making,albeit in a rudimentary form,could already be discerned in the writings of G.E.Moore and E.Husserl in the early 1900s.A few years later it was more extensively exploited by William James.It was Wittgenstein,however,who gave the concept a precise meaning.In 1913/1914 Wittgenstein advanced a theory of possible worlds,only one of which was real.Every proposition suggests a part of a possible world which does or does not correspond to parts of the real world.In the first case the proposition is true,in the second case false.Moreover,the part of the real world makes the sentence true.This part is a truth-maker,and the sentence is a truth-bearer.Surprisingly enough,Wittgenstein’s concept of truth-making had its family resemblance with William James’s conception of truth.In 1915 Wittgenstein stopped using the concept of truth-making-it was also not mentioned in the Tractatus.Unfortunately,Russell did not notice this and in 1918 he adopted the concept of truth-making.In the 1930s,it was used by some second generation analytic philosophers(Schlick,Stebbing,and Wisdom).However,it became rather popular among analytic philosophers only in the 1980s.
文摘Employing a 4-form ansatz of 11-dimensional supergravity over a non-dynamical AdS_(4)×S^(7)/Z_(k)background and setting the internal space as an S1 Hopf fibration on CP3,we obtain a consistent truncation.The(pseudo)scalars,in the resulting scalar equations in Euclidean AdS_(4)space,may be considered to arise from(anti)M-branes wrapping around the internal directions in the(Wick-rotated)skew-whiffed M2-brane background(as the resulting theory is for anti-M2-branes),thus realizing the modes after swapping the three fundamental representations 8_(s),8_(c),and 8_(v) of SO(8).Taking the backreaction on the external and internal spaces,we obtain the massless and massive modes,corresponding to exactly marginal and marginally irrelevant deformations on the boundary CFT3,respectively.Subsequently,we obtain a closed solution for the bulk equation and compute its correction with respect to the background action.Next,considering the Higgs-like(breathing)mode m^(2)=18,having all supersymmetries as well as parity and scale-invariance broken,solving the associated bulk equation with mathematical methods,specifically the Adomian decomposition method,and analyzing the behavior near the boundary of the solutions,we realize the boundary duals in the SU(4)×U(1)-singlet sectors of the ABJM model.Then,introducing the new dual deformationΔ_(+)=3,6 operators made of bi-fundamental scalars,fermions,and U(1)gauge fields,we obtain the SO(4)-invariant solutions as small instantons on a three-sphere with the radius at infinity,which correspond to collapsing bulk bubbles leading to big-crunch singularities.
