Electricity theft behavior has serious influence on the normal operation of power grid and economic benefits enterprises. Intelligent anti-power-theft algorithm is required for monitoring consumption data to recognize electricity theft. In this paper, an adaptive time-series recurrent neural network (TSRNN) architecture was built up detect abnormal users (i.e., users) in consumption. fusion with synthetic minority oversampling technique (SMOTE) algorithm, a batch virtual observations were...

High-quality gasoline combustion reduces atmospheric emission. The evaluation of quality is needed reduce air pollution. purity commonly quantitatively evaluated by octane number. Near-infrared (NIR) spectral determination the number practical for rapid quality. In NIR, calibration model must be optimized improvement accuracy using deep learning methodologies. this work, an algorithmic architecture was constructed hybrid optimization combination fractional derivative (FD) and partial least...

Library data contains many students’ reading records that reflect their general knowledge acquisition. The purpose of this study is to deeply mine the library book-borrowing data, with concerns on different book catalogues and properties predict extracurricular interests. An intelligent computing framework proposed by fusion a neural network architecture partial differential equations (PDE) function module. In model designs, constructed as an adaptive learning backpropagation (BPNN),...

Chemical oxygen demand (COD) is an important index to measure the degree of water pollution. In this paper, near-infrared technology used obtain 148 wastewater spectra predict COD value in wastewater. First, partial least squares regression (PLS) model was as basic model. Monte Carlo cross-validation (MCCV) select 25 samples out that did not conform conventional statistics. Then, interval (iPLS) modeling carried on 123 samples, and spectral bands were divided into 40 subintervals. The...

Pest detection is important for crop cultivation. Crop leaf the main place of pest invasion. Current technologies to detect pests have constraints, such as low efficiency, storage demands and limited precision. Image segmentation a fast efficient computer-aided technology. High resolution image capture solidly supports crucial processes in discerning from images. Study analytical methods help parse information In this paper, regional convolutional neural network (R-CNN) architecture designed...

“Low profit and high sales” is a strategy to increase sales volume by reducing the of unit goods, so that businesses can gain more profits. For flexible price reduction total revenue, but when goods are lack flexibility, will reduce revenue. In this paper, according data provided supermarket, we preprocess data, establish appropriate indicators measure daily discount strength mall, mathematical model between strength, margin. Through these models, found meager profits do bring up sales, too...

With the development of computer software and hardware system, machine learning methods are more used in various industries social development. In aspect stock index prediction, current prediction method has gradually changed from traditional statistical analysis to artificial intelligence method. Based on original sample data, this paper uses support vector regression (SVR) model predict opening price Shanghai index. The parameters SVR optimized debugged by grid search (grid), particle...

Investigation on college students’ consumption ability help classify them as from rich or relative poor family, thus to distinguish the students who are in urgent need for government’s economic support. As canteen is main part of expenses students, we proposed adjusted K-means clustering methods discrimination at different levels. To improve accuracy, a broad learning network architecture was built up extracting informative features records. A fuzzy transformed technique combined extend...

Optimal human resources allocation asks to employ a person work in the position corresponding his/her ability. Employment competence is key feedback cultivation of college students' working The data relationship needs analyze between in-school items and abilities required by companies. Machine learning framework introduced study companies' responses students. In this work, dual-network architecture built up for statistical modeling evaluation graduates' ability consistence with their job...

The quality of graduates is the key factor in evaluating cultivation effect colleges and universities. Quantification whether qualify for their working post companies industries provides conduction further college reform enhancement. In this work, we proposed an adaptive multivariate neural network architecture fusion evaluation student cultivation. Specifically, designed a questionnaire to collect data on current status 1231 recorded 32 in-school training items categorized into four...

With the rapid development of big data technology, personal credit evaluation industry has entered a new stage. Among them, based on mobile telecommunications is one hotspots current research. However, due to complexity and diversity variables, in order reduce model improve prediction accuracy model, we need dimension input variables. According provided by operator, this paper divides into training sets verification sets. We perform correlation analysis each indicator set, calculate...