The component loadings are interpreted by considering their magnitudes, which indicates how strongly each of the original variables relates to the corresponding principal component. The usual ad hoc practice in the interpretation process is to ignore the variables with small absolute loadings or set to zero loadings smaller than some threshold value. This, in fact, makes the component loadings sparse in an artificial and a subjective way. We propose a new alternative approach, which produces sparse loadings in an optimal way. The introduced approach is illustrated on two well-known data sets and compared to the existing rotation methods.

The factor analysis (FA) model does not permit unique estimation of the common and unique factor scores. This weakness is notorious as the factor indeterminacy in FA. Luckily, some part of the factor scores can be uniquely determined. Thus, as a whole, they can be viewed as a sum of determined and undetermined parts. The paper proposes to select the undetermined part, such that the resulting common factor scores have the following feature: the rows (i.e., individuals) of the common factor score matrix are as well classified as possible into few clusters. The clear benefit is that we can easily interpret the factor scores simply by focusing on the clusters. The procedure is called clustered common factor exploration (CCFE). An alternating least squares algorithm is developed for CCFE. It is illustrated with real data examples. The proposed approach can be viewed as a parallel to the rotation techniques in FA. They exploit another FA indeterminacy, the rotation indeterminacy, which is resolved by choosing the rotation that transforms the loading matrix into the ‘most’ interpretable one according to a pre-specified criterion. In contrast to the rotational indeterminacy, the factor indeterminacy is utilized to achieve well-clustered factor scores by CCFE. To the best of our knowledge, such an approach to the FA interpretation has not been studied yet.

Sparse principal component analysis is a very active research area in the last decade. It produces component loadings with many zero entries which facilitates their interpretation and helps avoid redundant variables. The classic factor analysis is another popular dimension reduction technique which shares similar interpretation problems and could greatly benefit from sparse solutions. Unfortunately, there are very few works considering sparse versions of the classic factor analysis. Our goal is to contribute further in this direction. We revisit the most popular procedures for exploratory factor analysis, maximum likelihood and least squares. Sparse factor loadings are obtained for them by, first, adopting a special reparameterization and, second, by introducing additional ${\ell }_1$\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$\ell _1$$\end{document}-norm penalties into the standard factor analysis problems. As a result, we propose sparse versions of the major factor analysis procedures. We illustrate the developed algorithms on well-known psychometric problems. Our sparse solutions are critically compared to ones obtained by other existing methods.

A new factor analysis (FA) procedure has recently been proposed which can be called matrix decomposition FA (MDFA). All FA model parameters (common and unique factors, loadings, and unique variances) are treated as fixed unknown matrices. Then, the MDFA model simply becomes a specific data matrix decomposition. The MDFA parameters are found by minimizing the discrepancy between the data and the MDFA model. Several algorithms have been developed and some properties have been discussed in the literature (notably by Stegeman in Comput Stat Data Anal 99:189–203, 2016), but, as a whole, MDFA has not been studied fully yet. A number of new properties are discovered in this paper, and some existing ones are derived more explicitly. The properties provided concern the uniqueness of results, covariances among common factors, unique factors, and residuals, and assessment of the degree of indeterminacy of common and unique factor scores. The properties are illustrated using a real data example.

Four microseconds long Ar3+ beam with injection energy of 15 keV/u has been injected into the Digital Accelerator of the High-Energy Accelerator Research Organization. Beam production, transportation, and injection are described as well as machine properties. Results of a free running experiment under static magnetic field and longitudinal confinement and acceleration under a fast ramping magnetic field are presented in detail with a brief discussion on the beam lifetime.

Extended abstract of a paper presented at Microscopy and Microanalysis 2008 in Albuquerque, New Mexico, USA, August 3 – August 7, 2008

To investigate two clusters of diarrhoea cases observed in our geriatric hospital wards, the faecal specimens were analysed. Reversed passive latex agglutination assay revealed that 63·2% and 41·7% of the faecal specimens from each cluster were positive for Clostridium perfringens enterotoxin. PCR assay revealed that 71·4% and 68·8% of C. perfringens isolates from each cluster were positive for the enterotoxin gene (cpe). These observations suggested that both the clusters were outbreaks caused by enterotoxigenic C. perfringens. Subsequent pulsed-field gel electrophoresis analysis revealed that the two outbreaks were caused by different C. perfringens isolates. However, these outbreak isolates as well as other sporadic diarrhoea isolates shared a 75-kb plasmid on which the cpe gene and the tcp locus were located. The 75-kb plasmid had horizontally spread to various C. perfringens isolates and had caused outbreaks and sporadic infections. However, the site and time of the plasmid transfer are unclear.

The propagation of intense laser pulses and the generation of high-energy electrons from underdense plasmas are investigated using two-dimensional particle-in-cell simulations. It is shown that quasi-monoenergetic electron beams are generated in the regime where the laser pulse length is much longer than the plasma wavelength, when the condition of the focusing is appropriately controlled.

Three-dimensional electron motion in a linearly polarized tightly focused laser field is numerically calculated. A high-intensity laser pulse focused on the free electrons in vacuum generates relativistic electron bunches whose length is shorter than the laser wavelength. The extremely short electron bunches with low-energy spread less than 1% are generated for a wide range of the laser parameters.

Positive surgical indications for an only hearing ear were evaluated in order to improve patients' quality of life. Fifteen cases of surgery involving an only hearing ear over the past eight years were retrospectively reviewed. Of eight perforated chronic otitis media cases, seven underwent type one tympanoplasty and one underwent simple underlay myringoplasty regardless of otorrhoea at the time of surgery. Of six cholesteatoma cases, two received the canal wall up method and four received the canal wall down method. Ossiculoplasty was carefully performed in six cases. Hearing was improved in seven cases, whereas it remained unchanged in seven cases and deteriorated in one case. Of nine patients, two did not need a hearing aid after surgery. Five patients with severe combined hearing loss (>90 dB) were able to communicate with a hearing aid, alleviating their anxiety regarding hearing loss. Only hearing ears with chronic otitis media and cholesteatoma can be successfully treated by tympanoplasty with or without ossiculoplasty.

We have demonstrated the acceleration of a monoenergetic electron beam by a laser-produced wakefield. Experiments were performed by focusing 2-TW laser pulses of 50 fs on supersonic gas-jet targets. The focused intensity was 5 × 1018 W/cm2 (a0 = 1.5). At an electron density of 1.5 × 1020 cm−3, the clear monoenergetic electron beam from the plasma was obtained at 7 to 15 MeV. The Stokes satellite peak in the forward scattering explained the energy spectra of electrons at various plasma densities well. Although the wakefield propagated 500 microns, which was far beyond the dephasing length, monoenergetic electron beams were obtained.

The Groningen voice prosthesis can be successfully replaced using the back-loading system. We have attempted to minimize patient stress by developing a front-loading system that does not require insertion of the introducer via the tracheo-oesophageal shunt to the oral cavity or the mesopharyngeal anaesthetization regularly used with the back-loading system. Using our front-loading system, the existing prosthesis is removed, then the posterior portion of the replacement Groningen prosthesis is grasped by a pair of nasal forceps with a small jaw to make an acute angle and inserted into the oesophageal cavity through the shunt at a stroke. All 20 patients who underwent forward-loading replacement of a Groningen (n=17) or Blom-Singer (n=3) valve with a Groningen valve tolerated the procedure well and experienced no complications except, in some cases, for minor bleeding just after insertion of the prosthesis.The procedure was completed within 30 seconds.

Active matrix organic light emitting diode displays based on thin, flexible metal foil substrates offer a novel approach to fabricate light, flexible and rugged displays. Metal foils allows devices to be fabricated at higher process temperatures yielding better device characteristics than plastic substrates. Furthermore, the conductive nature of such substrates enables more efficient use of pixel area through the use of a global power electrode. We are implementing Active Matrix Organic Light Emitting Diode (AMOLED) displays on flexible stainless steel substrate using 2 TFT and 4 TFT pixel circuit topologies with two different layout realizations that exploit the conductive properties of the substrate.

We report that polycrystalline cubic-Si3N4 with a spinel structure and low oxygen concentration (<0.5 wt%) shows Vickers hardness of 43 GPa when measured with the indentation load of 10 mN. The hardness decreases with the increase of the indentation load, which can be ascribed to the presence of weak grain boundaries. The high hardness can be well explained by its large shear modulus as predicted by first-principles calculations.

The dielectric properties of the Ba(Ti1–xZrx)O3 (BT1–xZx) ceramics with various compositions were investigated under various DC bias fields. In BT0.94Z0.06 ceramics, the dielectric constants (ε) vs temperature (T) curves without DC bias showed no frequency dependence. Its DC bias dependence exhibited that with increasing DC bias fields, the dielectric constants were suppressed while phase transition peaks slightly shifted to high temperature. In BT0.58Z0.42 ceramics, the ε vs T curves without DC bias showed a broad peak with a clear frequency dependence, which revealed that BT0.58Z0.42 ceramics were the relaxor. Its DC bias dependence exhibited that with increasing DC bias fields, the dielectric constants were suppressed while the phase transition peak largely shifted to low temperatures. In BT0.79Z0.21 ceramics, the ε vs T curves without DC bias showed a broad peak without frequency dependence. Its DC bias dependence revealed that the dielectric peak shifted to high temperature and broadened with increasing DC bias. To explain the above phenomena, it was considered that the role of Zr ions on BT1−xZx ceramics is to make the depth of the potential well shallow successively.