Context-Aware Approach for Restaurant Recommender Systems

2018 ◽  
pp. 1757-1771
Author(s):  
Haoxian Feng ◽  
Thomas Tran
Keyword(s):  
Root Mean Square Error ◽  
Collaborative Filtering ◽  
Recommender Systems ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Mean Square ◽  
Recommendation Algorithm ◽  
Filtering Method ◽  
User Behaviors ◽  
Svd Method

This paper addresses the issue of how to effectively use users' historical data in restaurant recommender systems, as opposed to systems, such as FindMe, that only rely on online operations. Towards that end, the authors propose a bias-based SVD method as the underlying recommendation algorithm and test it against the traditional item-based collaborative filtering method on the Entrée restaurant dataset. The results are promising as the obtained Root-Mean-Square-Error (RMSE) values reach 0.58 for the SVD and 0.62 for the item-based system. Researchers can extend the transformation from user behaviors to ratings in more application domains other than the restaurant one.

Context-Aware Approach for Restaurant Recommender Systems

2019 ◽  
pp. 324-341 ◽  
Author(s):  
Haoxian Feng ◽  
Thomas Tran
Keyword(s):  
Root Mean Square Error ◽  
Collaborative Filtering ◽  
Recommender Systems ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Mean Square ◽  
Recommendation Algorithm ◽  
Filtering Method ◽  
User Behaviors ◽  
Svd Method

This chapter addresses the issue of how to effectively use users' historical data in restaurant recommender systems, as opposed to systems, such as FindMe, that only rely on online operations. Towards that end, the authors propose a bias-based SVD method as the underlying recommendation algorithm and test it against the traditional item-based collaborative filtering method on the Entrée restaurant dataset. The results are promising as the obtained root-mean-square-error (RMSE) values reach 0.58 for the SVD and 0.62 for the item-based system. Researchers can extend the transformation from user behaviors to ratings in more application domains other than the restaurant one.

scholarly journals Ανάπτυξη πολυκριτήριου συστήματος συστάσεων για προσωποποιημένο μάρκετινγκ

2018 ◽  
Author(s):  
Ιωάννης Ζήσος
Keyword(s):  
Root Mean Square Error ◽  
Recommender Systems ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Mean Absolute Error ◽  
Cold Start ◽  
Absolute Error ◽  
Percentage Error ◽  
Mean Square ◽  
Absolute Percentage Error

Ένα από τα πιο καίρια ερωτήματα για τους χρήστες του διαδικτύου, είναι πώς θα καταφέρουν να διαχειριστούν την τεράστια ποσότητα διαθέσιμης πληροφορίας, ώστε να καταλήξουν σε επιλογή προϊόντων που ανταποκρίνονται όσο το δυνατό καλύτερα στις προτιμήσεις και ανάγκες τους. Αντίστοιχα, οι εταιρείες που παρέχουν προϊόντα ή υπηρεσίες μέσω του διαδικτύου, προσπαθούν συστηματικά να εντοπίσουν μεθόδους ώστε να αποκωδικοποιήσουν με ακρίβεια τα προφίλ προτίμησης των χρηστών, με στόχο να καταφέρουν να προσαρμόσουν κατάλληλα τα προϊόντα τους και να αυξήσουν τις πωλήσεις τους.Για τους παραπάνω λόγους, η επιστημονική και ερευνητική κοινότητα που δραστηριοποιείται στο τομέα της ανάλυσης δεδομένων και το μάρκετινγκ έχει επικεντρώσει την προσπάθεια της, στην δημιουργία μεθοδολογιών που θα απαντήσουν όσο το δυνατόν πιο αποτελεσματικά τα παραπάνω ερωτήματα. Οι περισσότερες από αυτές τις μεθοδολογίες καταλήγουν στην ανάπτυξη προσαρμοστικών συστημάτων που αντλούν δεδομένα από το διαδίκτυο και εξάγουν προτάσεις για τους χρήστες. Η πιο γνωστή κατηγορία τέτοιου είδους συστημάτων είναι τα συστήματα συστάσεων (Recommender Systems). Στην παρούσα ερευνητική εργασία παρουσιάζεται η μεθοδολογία και τα αποτελέσματα πιλοτικής λειτουργίας ενός νέου υβριδικού συστήματος συστάσεων που βασίζεται στη χρήση μεθόδων ανάλυσης συναισθήματος, πολυκριτήριας ανάλυσης καθώς και μεθόδων φιλτραρίσματος. Η μεθοδολογία καταλήγει σε τέσσερα διαφορετικά είδη σύστασης, με άκρως ενδιαφέροντα αποτελέσματα.Μέσω του μεθοδολογικού πλαισίου γίνεται εφικτός ο προσδιορισμός των προτιμησιακών προφίλ των χρηστών του συστήματος, τα οποία εν συνεχεία αντιστοιχίζονται σε «προφίλ πελατών» που επιλέγουν συγκεκριμένα προϊόντα/υπηρεσίες που τους «ταιριάζουν».Έτσι, καταλήγουμε σε προσωποποιημένες συστάσεις προϊόντων στον χρήστη του συστήματος, που είναι ανάλογες των προτιμήσεων του. Επιπλέον δίνεται στο χρήστη η δυνατότητα να φιλτράρει τις διαθέσιμες εναλλακτικές με σχετική επιλογή από ένα σύνολο κατ’ αποκοπή κριτηρίων. Η χρήση του κατωφλιού ελάχιστης ικανοποίησης, που προσδιορίζεται από τα αποτελέσματα της ανάλυσης συναισθήματος στα σχόλια των πελατών, εγγυάται την ποιότητα των συστάσεων.Τα δεδομένα του συστήματος είναι πραγματικές απόψεις και βαθμολογίες χρηστών για καταλύματα, καθώς και χαρακτηριστικά καταλυμάτων που αντλήθηκαν από γνωστή διαδικτυακή πλατφόρμα κρατήσεων. H ανάπτυξη του συστήματος βασίστηκε στην μεθοδολογία CRISP-DM(Shearer, 2000a). Η αξιολόγηση του συστήματος συστάσεων γίνεται με μέτρηση της ακρίβειας προβλέψεων αξιολογήσεων σε πείραμα με πραγματικούς χρήστες.Για τη μελέτη περίπτωσης χρησιμοποιήθηκαν δεδομένα για τα τουριστικά καταλύματα του Νομού Χανίων. Τα πιο σημαντικά χαρακτηριστικά της συγκεκριμένης ερευνητικής προσπάθειας είναι: α. H χρήση μεγάλου όγκου πραγματικών δεδομένων σε αντίθεση με τις περισσότερες από τις υπάρχουσες έρευνες που χρησιμοποιούν έτοιμα τεστ σετ δεδομένων. β. Η χρησιμοποίηση όλης της διαθέσιμης αντλημένης πληροφορίας προκειμένου να καταλήξουμε σε σύσταση. Πιο συγκεκριμένα στην παρούσα μεθοδολογία χρησιμοποιούνται δεδομένα βαθμολογιών προϊόντων για την ανάλυση ικανοποίησης πελατών, δεδομένα απόψεων για τη συναισθηματική ανάλυση, στατικά δεδομένα των προϊόντων σαν κατ’ αποκοπή κριτήρια σε αντίθεση με τις περισσότερες μελέτες όπου χρησιμοποιούνται είτε βαθμολογίες, είτε ανάλυση σχολίων για την τελική σύσταση.γ. Η χρήση των αποτελεσμάτων της ανάλυσης συναισθήματος ως κατώφλια ποιότητας σύστασης. δ. Η επιτυχής εφαρμογή παραλλαγής της πρόσφατα παρουσιασθείσας μεθόδου WAP, για τη δημιουργία προτιμησιακού προφίλ χρήστη.ε. Η απαίτηση για εισαγωγή ελάχιστων δεδομένων από το χρήστη.στ. Ο ελάχιστος χρόνος που απαιτείται για την παραγωγή της σύστασης κατά τη λειτουργία του συστήματος.η. Η αποφυγή άντλησης προσωπικών δεδομένων του χρήστη για χρήση τους στη διαδικασία της σύστασης. θ. H αποφυγή του προβλήματος της καθυστερημένης εκκίνησης (cold start). ι. Τα άκρως ικανοποιητικά αποτελέσματα με βάση τις μετρικές που χρησιμοποιήθηκαν για τη δοκιμή του συστήματος.κ. Η χρησιμότητα του συστήματος είναι χαρακτηριστική για όλες τις ομάδες στόχου, καθώς δίνεται η δυνατότητα παροχής χρήσιμης πληροφορίας τόσο στον πελάτη με σύσταση ανάλογη των αναγκών του, όσο και στον πάροχο υπηρεσίας/προϊόντος, προσδιορίζοντας του τις τάσεις όσον αφορά την ικανοποίηση των πελατών. Οι κύριες μετρικές που χρησιμοποιούνται για την μέτρηση της ποιότητας των συστάσεων που παράγονται είναι οι: Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE).H διατριβή ολοκληρώνεται, με προτάσεις, για μελλοντική έρευνα και επέκταση του παρόντος συστήματος.

scholarly journals Estimation of Apple Flowering Frost Loss for Fruit Yield Based on Gridded Meteorological and Remote Sensing Data in Luochuan, Shaanxi Province, China

2021 ◽  
Vol 13 (9) ◽  
pp. 1630
Author(s):  
Yaohui Zhu ◽  
Guijun Yang ◽  
Hao Yang ◽  
Fa Zhao ◽  
Shaoyu Han ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Minimum Temperature ◽  
Price Regulation ◽  
Flowering Period ◽  
Mean Square ◽  
Prevention Measures ◽  
Agricultural Insurance

With the increase in the frequency of extreme weather events in recent years, apple growing areas in the Loess Plateau frequently encounter frost during flowering. Accurately assessing the frost loss in orchards during the flowering period is of great significance for optimizing disaster prevention measures, market apple price regulation, agricultural insurance, and government subsidy programs. The previous research on orchard frost disasters is mainly focused on early risk warning. Therefore, to effectively quantify orchard frost loss, this paper proposes a frost loss assessment model constructed using meteorological and remote sensing information and applies this model to the regional-scale assessment of orchard fruit loss after frost. As an example, this article examines a frost event that occurred during the apple flowering period in Luochuan County, Northwestern China, on 17 April 2020. A multivariable linear regression (MLR) model was constructed based on the orchard planting years, the number of flowering days, and the chill accumulation before frost, as well as the minimum temperature and daily temperature difference on the day of frost. Then, the model simulation accuracy was verified using the leave-one-out cross-validation (LOOCV) method, and the coefficient of determination (R2), the root mean square error (RMSE), and the normalized root mean square error (NRMSE) were 0.69, 18.76%, and 18.76%, respectively. Additionally, the extended Fourier amplitude sensitivity test (EFAST) method was used for the sensitivity analysis of the model parameters. The results show that the simulated apple orchard fruit number reduction ratio is highly sensitive to the minimum temperature on the day of frost, and the chill accumulation and planting years before the frost, with sensitivity values of ≥0.74, ≥0.25, and ≥0.15, respectively. This research can not only assist governments in optimizing traditional orchard frost prevention measures and market price regulation but can also provide a reference for agricultural insurance companies to formulate plans for compensation after frost.

scholarly journals Low Cost Automatic Reconstruction of Tree Structure by AdQSM with Terrestrial Close-Range Photogrammetry

Forests ◽  
2021 ◽  
Vol 12 (8) ◽  
pp. 1020
Author(s):  
Yanqi Dong ◽  
Guangpeng Fan ◽  
Zhiwu Zhou ◽  
Jincheng Liu ◽  
Yongguo Wang ◽  
...  
Keyword(s):  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Low Cost ◽  
Reference Value ◽  
Tree Height ◽  
Point Clouds ◽  
Tree Structure ◽  
Mean Square ◽  
Tree Model

The quantitative structure model (QSM) contains the branch geometry and attributes of the tree. AdQSM is a new, accurate, and detailed tree QSM. In this paper, an automatic modeling method based on AdQSM is developed, and a low-cost technical scheme of tree structure modeling is provided, so that AdQSM can be freely used by more people. First, we used two digital cameras to collect two-dimensional (2D) photos of trees and generated three-dimensional (3D) point clouds of plot and segmented individual tree from the plot point clouds. Then a new QSM-AdQSM was used to construct tree model from point clouds of 44 trees. Finally, to verify the effectiveness of our method, the diameter at breast height (DBH), tree height, and trunk volume were derived from the reconstructed tree model. These parameters extracted from AdQSM were compared with the reference values from forest inventory. For the DBH, the relative bias (rBias), root mean square error (RMSE), and coefficient of variation of root mean square error (rRMSE) were 4.26%, 1.93 cm, and 6.60%. For the tree height, the rBias, RMSE, and rRMSE were—10.86%, 1.67 m, and 12.34%. The determination coefficient (R2) of DBH and tree height estimated by AdQSM and the reference value were 0.94 and 0.86. We used the trunk volume calculated by the allometric equation as a reference value to test the accuracy of AdQSM. The trunk volume was estimated based on AdQSM, and its bias was 0.07066 m3, rBias was 18.73%, RMSE was 0.12369 m3, rRMSE was 32.78%. To better evaluate the accuracy of QSM’s reconstruction of the trunk volume, we compared AdQSM and TreeQSM in the same dataset. The bias of the trunk volume estimated based on TreeQSM was −0.05071 m3, and the rBias was −13.44%, RMSE was 0.13267 m3, rRMSE was 35.16%. At 95% confidence interval level, the concordance correlation coefficient (CCC = 0.77) of the agreement between the estimated tree trunk volume of AdQSM and the reference value was greater than that of TreeQSM (CCC = 0.60). The significance of this research is as follows: (1) The automatic modeling method based on AdQSM is developed, which expands the application scope of AdQSM; (2) provide low-cost photogrammetric point cloud as the input data of AdQSM; (3) explore the potential of AdQSM to reconstruct forest terrestrial photogrammetric point clouds.

Comparison of Thresholding and Edge Detection Segmentation Techniques

2013 ◽  
Vol 860-863 ◽  
pp. 2783-2786
Author(s):  
Yu Bing Dong ◽  
Hai Yan Wang ◽  
Ming Jing Li
Keyword(s):  
Image Segmentation ◽  
Root Mean Square Error ◽  
Edge Detection ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Experimental Results ◽  
Mean Square ◽  
Segmentation Method ◽  
Segmentation Algorithms

Edge detection and thresholding segmentation algorithms are presented and tested with variety of grayscale images in different fields. In order to analyze and evaluate the quality of image segmentation, Root Mean Square Error is used. The smaller error value is, the better image segmentation effect is. The experimental results show that a segmentation method is not suitable for all images segmentation.

Quantitative Analysis of Forsythiae fructus by Near Infrared Spectroscopy

2013 ◽  
Vol 807-809 ◽  
pp. 1967-1971
Author(s):  
Yan Bai ◽  
Xiao Yan Duan ◽  
Hai Yan Gong ◽  
Cai Xia Xie ◽  
Zhi Hong Chen ◽  
...  
Keyword(s):  
Infrared Spectroscopy ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Cross Validation ◽  
Ethanol Extract ◽  
Mean Square ◽  
Validation Set

In this paper, the content of forsythoside A and ethanol-extract were rapidly determinated by near-infrared reflectance spectroscopy (NIRS). 85 samples of Forsythiae Fructus harvested in Luoyang from July to September in 2012 were divided into a calibration set (75 samples) and a validation set (10 samples). In combination with the partical least square (PLS), the quantitative calibration models of forsythoside A and ethanol-extract were established. The correlation coefficient of cross-validation (R2) was 0.98247 and 0.97214 for forsythoside A and ethanol-extract, the root-mean-square error of calibration (RMSEC) was 0.184 and 0.570, the root-mean-square error of cross-validation (RMSECV) was 0.81736 and 0.36656. The validation set were used to evaluate the performance of the models, the root-mean-square error of prediction (RMSEP) was 0.221 and 0.518. The results indicated that it was feasible to determine the content of forsythoside A and ethanol-extract in Forsythiae Fructus by near-infrared spectroscopy.

Analysis of candlenut oil as oil adulterant in three functional oils of soybean oil, sunflower oil and grapeseed oil in quaternary mixture systems using FTIR spectroscopy and chemometrics

Food Research ◽  
2021 ◽  
Vol 5 (2) ◽  
pp. 248-253
Author(s):  
A.B. Riyanta ◽  
S. Riyanto ◽  
E. Lukitaningsih ◽  
A. Rohman
Keyword(s):  
Root Mean Square Error ◽  
Soybean Oil ◽  
Ftir Spectroscopy ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Sunflower Oil ◽  
Ftir Spectra ◽  
Mean Square ◽  
Grapeseed Oil ◽  
Quaternary Mixture

Soybean oil (SBO), sunflower oil (SFO) and grapeseed oil (GPO) contain high levels of unsaturated fats that are good for health and have proximity to candlenut oil. Candlenut oil (CNO) has a lower price and easier to get oil from that seeds than other seed oils, so it is used as adulteration for gains. Therefore, authentication is required to ensure the purity of oils by proper analysis. This research was aimed to highlight the FTIR spectroscopy application with multivariate calibration is a potential analysis for scanning the quaternary mixture of CNO, SBO, SFO and GPO. CNO quantification was performed using multivariate calibrations of principle component (PCR) regression and partial least (PLS) square to predict the model from the optimization FTIR spectra regions. The highest R2 and the lowest values of root mean square error of calibration (RMSEC) and root mean square error of prediction (RMSEP) were used as the basis for selection of multivariate calibrations created using several wavenumbers region of FTIR spectra. Wavenumbers regions of 4000-650 cm-1 from the second derivative FTIR-ATR spectra using PLS was used for quantitative analysis of CNO in quaternary mixture with SBO, SFO and GPO with R2 calibration = 0.9942 and 0.0239% for RMSEC value and 0.0495%. So, it can be concluded the use of FTIR spectra combination with PLS is accurate to detect quaternary mixtures of CNO, SBO, SFO and GPO with the highest R2 values and the lowest RMSEC and RMSEP values.

scholarly journals A useful quantitative model for determination of enantiomeric composition of racemate praziquantel by ultraviolet spectroscopy combined with partial least squares and its application to praziquantel tablets

2018 ◽  
Vol 11 (03) ◽  
pp. 1850011 ◽  
Author(s):  
Man Zhao ◽  
Ran Meng ◽  
Yifang Lu ◽  
Lingyun Hu ◽  
Na Sun ◽  
...  
Keyword(s):  
Root Mean Square Error ◽  
Least Squares ◽  
Partial Least Squares ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Reference Value ◽  
Quantitative Model ◽  
Enantiomeric Composition ◽  
Mean Square ◽  
Validation Set

A simple and novel method has been proposed to determine the enantiomeric composition of racemate praziquantel (PZQ) by using the analysis of ultraviolet (UV) spectroscopy combined with partial least squares (PLS). This method does not rely on the use of expensive carbohydrates such as cyclodextrins, but on the use of inexpensive sucrose, which is equally effective as carbohydrate. PZQ has two enantiomers. Through measuring the slight difference in the UV spectral absorption of PZQ due to different interactions between its two enantiomers and sucrose, the enantiomeric composition was determined by a quantitative model based on PLS analysis. The model showed that the correlation coefficients of calibration set and validation set were 0.9971 and 0.9972, respectively. The root mean square error of calibration (RMSEC) and the root mean square error of prediction (RMSEP) were 0.0167 and 0.0129, respectively. Then, the independent data of PZQ tablets were also used to test how well the quantitative model of PLS predicted the enantiomeric composition. The ratio of S-PZQ in tablet was 0.492, determined by high-performance liquid chromatography as the reference value. Six solutions of the tablet samples were prepared, and the ratios of S-PZQ in tablet samples in the validation set were predicted by the PLS model. Their relative errors with the reference value were not more than 4%. Therefore, the established model could be accurate and employed to predict the enantiomeric compositions of PZQ tablets.

Sign in / Sign up

Export Citation Format

Share Document