Search results

Experimental data sets prove that the proposed algorithm is effective and reasonable. 1.
In such a large amount of data and sparse, on the one hand, difficult to find the nearest neighbor sets of users, on the other hand the cost of similarity calculations will be great, is shown by equation 2
(3) To better solve the data sparseness of collaborative filtering recommender system implementations, synonyms (similar products with a different name to describe and can not find this correlation), and the present use is widely used in information retrieval dimensionality reduction - latent semantic indexing (Latent the Semantic the Indexing, LSI), used to solve the problem of synonyms and policemen.
But do not need the user's historical data, or user identification.
The model-based collaborative filtering (Model-based Collaborative Filtering) is to use historical data to a model, then this model to predict.

All measurement processes produce results for which the appropriate data analysis is performed.
Analysis " PG-05-07-ANX-1 Data processing techniques To calculate the costs of execution of an activity, general information concerning the costs of the recoded activity is appended to these records.
When direct data allowing to calculate the probability of a failure mode are not available or this failure form is complex, it is proposed the elaboration of a fault tree.
Its probability is calculated based on the probabilities of the basic events and undeveloped events to be evaluated from available historical data.
Costs data of the integrated management system are linked to a worksheet in which are kept current total costs of implementation of all activities under the integrated management system.

The latest results consisting in the development of a new algorithm, providing a more reliable evaluation of the test data will be described.
Thus, it was worth to execute the noise reduction on the edge image.
These values are based on some very first calculations, thus detailed numerical data are not yet presented.

Using statistical data from year 2003 to 2011, this paper computes the location quotient and cluster scale of Jiangsu textiles industry, which shows a remarkable cluster tendency.
Whereas data in Gross Industrial Output Value and “delivery value for export” were calculated on the basis of textiles, clothing and other fiber products manufacturing; leather, fur, down and its products; and chemical fiber manufacturing.
It can be seen from Fig.1 (derived from Table 1 data) that textiles delivery value for export of Jiangsu kept increasing except in 2009 which reflected the impact of global economic crisis.
Table 2 is derived by adding up data of textiles, clothing and fiber products manufacturing, leather, fur, down and its products, and chemical fiber manufacturing during 1998～2011.
Whereas data in Gross Industrial Output Value and “delivery value for export” were calculated on the basis of textiles, clothing and other fiber products manufacturing; leather, fur, down and its products; and chemical fiber manufacturing.

Samples were rolled at the laboratory rolling mill between plates of stainless steel up to the deformation of εmax= 80% with ~5% reductions per pass.
The treatment of texture data included construction of diagrams [PF(0001)recr - PF(0001)roll], where difference between pole densities for recrystallization and rolling textures are calculated at each point (ψ, φ), so that regions of maximal texture changes are seen.
Discussion First of all, obtained experimental results contain much new data, concerning the role of deformation twinning in recrystallization processes in α-Zr.

Experimental data of skim milk drying was used to validate the simulator.
Linear variation of Tgi and its influence on Xo Based on data shown in Fig 2., an exponential regression model was obtained with a coefficient of determination, R2, equal to 0.9963.
Therefore, the moisture will not evaporate fully because of reduction in drying time, increasing then, Xo, Yo and RHo.

Reduction in Defect Rate by Taguchi Method in Plastic Injection Molded Components G.
A training data set was created and 21 runs were carried out for training.

It shows that the predictions of the models are in good agreement with the experimental data.
To consider this effect, a proportionality factor, k, is introduced which can be determined from experimental data.
Figs.1 and 2 show the predicted trends (in lines) and experimental data (in symbols) for straight and contour cutting, respectively.
It is encouraging that the predicted trends are in good agreement with the experimental data.
Quantitative comparisons were made based on the percentage deviation of the predicted results from experimental data, as shown in Fig.3.

The data from GPC measure demonstrates that POE chains degraded during grafting process.
It is well known that ethylene units undergo predominantly crosslinking reactions (through the secondary macroalkyl radicals) during melt processing (see Scheme 1, b and c) resulting in an increase in molar mass and viscosity of the polymer, whereas octene units undergo primarily β-chain scission (via the tertiary macroalkyl radicals; see Scheme 1, d), leading to reduction in the molar mass[21].
The data from GPC measure demonstrates POE chains degraded during graft copolymerization process.

Finite element analysis was carried out and the result was in good agreement with the experimental data.
It has been demonstrated that when adhesively bonded straps or stingers had been used in box beams there is not crack arrest but a reduction of crack tip stress intensity, DK, that consequently produces a fatigue crack growth retardation effect [4-5].
The Harter T method, where the Walker equation, is used on a point-by-points basis, was used to interpolate the data between the R-ratios which were experimentally determined by equation (6), results shown in Fig.4