Attribute reduction is one of the most important preprocessing steps in machine learning and data mining. As a key step of attribute reduction, attribute evaluation directly affects classification ...performance, search time, and stopping criterion. The existing evaluation functions are greatly dependent on the relationship between objects, which makes its computational time and space more costly. To solve this problem, we propose a novel separability-based evaluation function and reduction method by using the relationship between objects and decision categories directly. The degree of aggregation (DA) of intraclass objects and the degree of dispersion (DD) of between-class objects are first defined to measure the significance of an attribute subset. Then, the separability of attribute subsets is defined by DA and DD in fuzzy decision systems, and we design a sequentially forward selection based on the separability (SFSS) algorithm to select attributes. Furthermore, a postpruning strategy is introduced to prevent overfitting and determine a termination parameter. Finally, the SFSS algorithm is compared with some typical reduction algorithms using some public datasets from UCI and ELVIRA Biomedical repositories. The interpretability of SFSS is directly presented by the performance on MNIST handwritten digits. The experimental comparisons show that SFSS is fast and robust, which has higher classification accuracy and compression ratio, with extremely low computational time.
We show that the class of C-hereditarily conjugacy separable groups is closed under taking arbitrary graph products whenever the class C is an extension closed variety of finite groups. As a ...consequence we show that the class of C-conjugacy separable groups is closed under taking arbitrary graph products. In particular, we show that right angled Coxeter groups are hereditarily conjugacy separable and 2-hereditarily conjugacy separable, and we show that infinitely generated right angled Artin groups are hereditarily conjugacy separable and p-hereditarily conjugacy separable for every prime number p.
Display omitted
Uniform Bi2MoO6 nanosheets were grown in a high dispersed fashion on electrospun BiFeO3 nanofibers via a solvothermal technique. The loading amount of Bi2MoO6 in the Bi2MoO6/BiFeO3 ...heterojunction nanofibers could be controlled by adjusting the precursor concentrations in the solvothermal process. The XPS analysis, energy band position calculation and trapping experiments all proved that the Bi2MoO6/BiFeO3 heterojunction is a Z-scheme heterojunction. The Z-scheme Bi2MoO6/BiFeO3 heterojunction had a much higher photocatalytic activity in the visible-light photodegradation of Rhodamine B (RhB) and tetracycline hydrochloride (TC) than pure BiFeO3 nanofibers or pure Bi2MoO6 nanosheets. The enhanced photocatalytic activity was attributed to the formation of Z-scheme Bi2MoO6/BiFeO3 heterojunctions, which could be beneficial to the separation of photogenerated electron-hole pairs. Moreover, the Bi2MoO6/BiFeO3 heterojunction nanofibers could be easily separated under an external magnetic field via the ferromagnetic BiFeO3. After several cycles, the photocatalytic activity of the Bi2MoO6/BiFeO3 heterojunction no longer significantly decreased suggesting that the Bi2MoO6/BiFeO3 heterojunction is stable. These Z-scheme Bi2MoO6/BiFeO3 heterojunction nanofibers with highly visible-light photocatalytic activity, excellent chemical stability and magnetic separability could be useful in many practical applications.
In this paper, we construct a separable reversible data hiding scheme for encrypted JPEG bitstreams. Our proposed scheme is constructed via a reserving-room-before-encryption manner, that is, the ...original JPEG bitstream is modified with small distortion so that the content owner can reserve enough space for future data embedding and then the modified JPEG bitstream is encrypted. To do that, our key observation is that the least significant bits of the two-bit appended values in the JPEG bitstream is a biased bitstream. Thus we design a new lossless compression algorithm for biased bitstreams with better compression ratio than the binary arithmetic coding method to fulfill the above task for pre-reserving space. With this room-reserving technique, the encrypted JPEG bitstream can be generated and the data embedding in this encrypted bitstream can be done easily by the data hider. Finally, the receiver is able to extract the embedded data and recovers the original JPEG bitstream independently.
•A separable reversible data hiding scheme for encrypted JPEG bitstreams is proposed.•The encryption method keeps the JPEG file structure unchanged.•A new lossless compression algorithm for biased bitstreams is proposed.•The proposed scheme is constructed via a reserving-room-before-encryption manner.
In this work, we present a study about the non-separability of degrees of freedom (DoF) of light in mixed modes, which emulates entangled mixed states. We explore bipartite spin–orbit modes and, by ...adding path DoF to spin–orbit modes, we propose an optical circuit to prepare tripartite non-separable mixed modes that extend the scenario of the classical-quantum analogy of non-separable modes. By using Machine Learning it was possible to show that we can characterize non-separability unambiguously with partial tomography measurements only on Sz basis, which is a remarkable advantage for higher dimension modes avoiding many tomographic measurements.
•Separability verification of mixed modes by partial tomography and machine learning.•Linear optical circuit for tripartite non-separable mixed optical modes preparation.•Linear optical circuit for tomography of tripartite mixed optical modes.•Emulation of entangled tripartite mixed states with intense laser beam.
Consider the fundamental group
of an arbitrary graph of groups and some root class
of groups, i.e., a class containing a nontrivial group and closed under subgroups, extensions, and unrestricted ...direct products of the form
, where
and
is an isomorphic copy of
for each
. We provide some criterion for the separability by
of a finitely generated abelian subgroup of
valid when the group satisfies an analog of the Baumslag filtration condition. This enables us to describe the
-separable finitely generated abelian subgroups for the fundamental groups of some graphs of groups with central edge subgroups.
Conventional magnetic biochar presents disadvantages such as fragility and high cost of imparting magnetic properties, and its fragility leads to the risk of free dispersion and low recovery rate in ...field applications. To solve these problems, sulfoaluminate cement was used as cementitious material, mixed with natural magnetite and biochar, and combined with foaming technology to prepare a new granular adsorbent in this work. The adsorbent exhibited excellent properties with compressive strength (1000 mN), specific surface area (49.32 m2 g−1), and magnetic properties (13.95 emu g−1). The adsorption capacity for Cd2+ was 45.94 mg g−1 at 25 °C, and it increased with the increase of ambient temperature. The adsorption mechanism was mainly dominated by the chemical interactions of monolayer adsorption, including coordination and complexation between Cd2+ and surface functional groups (e.g. –COO–Cd, –O–Cd), Cd–π interactions on aromatic structures, and the precipitation mechanism (Cd(OH)2, CdCO3).
Display omitted
•Using FMSB to adsorb cadmium realizes the use of waste to treat waste.•FMSB with certain compressive strength solves the fragility of biochar.•The use of magnetite as the magnetic source reduces the cost of imparting magnetism.•Foaming technology facilitates the mass transfer of cadmium to the interior of FMSB.•FMSB is safe in application as it does not leach hazardous substances in solution.
In this paper, we study the nonnegative matrix factorization problem under the separability assumption (that is, there exists a cone spanned by a small subset of the columns of the input nonnegative ...data matrix containing all columns), which is equivalent to the hyperspectral unmixing problem under the linear mixing model and the pure-pixel assumption. We present a family of fast recursive algorithms and prove they are robust under any small perturbations of the input data matrix. This family generalizes several existing hyperspectral unmixing algorithms and hence provides for the first time a theoretical justification of their better practical performance.