Low-level Data Fusion Research Articles

Species discrimination and total polyphenol prediction of porcini mushrooms by fourier transform mid-infrared (FT-MIR) spectrometry combined with multivariate statistical analysis.

The plateau specialty agricultural products, wild porcini mushrooms, have great value both as a superb cuisine and as a potential medication. Due to quality different between species added with the fraud behavior in sales process, make poor quality or poisonous sample inflow into the market, which pose a health risk for consumers, but also disrupted the mushroom market. Traditional analysis way is time‐consuming and laborious. Therefore, the aim of this study is to develop a way using fourier transform mid‐infrared (FT‐MIR) spectrometry and data fusion strategies for the fast and accurate species discrimination and predict amount of total polyphenol in four porcini mushrooms. The t‐distributed stochastic neighbor embedding based on mid‐level data fusion showed two species of Boletus edulis and B. umbriniporus have been identified. The order of correct rate of PLS‐DA models was mid‐level data fusionq (100%) > mid‐level data fusione (97.06%) = mid‐level data fusionv (97.06%) = stipes (97.06%) > low‐level data fusion (94.12%) > caps (91.18%). The order of correct rate of grid‐search support vector machine models was low‐level data fusion (100%) > caps (94.12%) > stipes (91.18%), and the order of particle swarm optimization support vector machine was low‐level data fusion (100%) > caps (97.06%) > stipes (88.24%). The mid‐level data fusionq and low‐level data fusion had best discrimination accuracy (100%) allowing each mushroom classed into its real species, which could be used for accurate discrimination of samples. B. edulis mushrooms had highest total polyphenol, with 14.76 mg/g dw and 17.33 in caps and stipes mg/g dw, respectively. The phenols were easier to accumulate in the caps in Leccinum rugosiceps (1.03) and B. tomentipes (1.19), and the opposite phenomenon is observed in B. edulis (0.85) and B. umbriniporus (0.95). The correlation coefficient and residual predictive deviation of best prediction model were 86.76% and 2.40%, respectively, indicating that that there is good relevance between FT‐MIR and total polyphenol content, which could be used to predict roughly polyphenols content in mushrooms.

Low-Level Fusion of Fourier Transform Ion Cyclotron Resonance Mass Spectrometry Data Sets for the Characterization of Nitrogen and Sulfur Compounds in Vacuum Gas Oils.

A total of 18 vacuum gas oils have been analyzed by Fourier transform ion cyclotron resonance mass spectrometry considering six replicates in three different ionization modes (electrospray ionization (ESI)(+), ESI(-), and atmospheric pressure photoionization (APPI)(+)) to characterize the nitrogen and sulfur compounds contained in these samples. Classical data analysis has been first performed on generated data sets using double bond equivalents (DBE) versus number of carbon atoms (#C) plots in order to observe similarities and differences within the nitrogen and sulfur-containing molecular classes from samples produced by different industrial processes. In a second step, three-way arrays have been generated for each ionization mode considering three dimensions: DBE related to aromaticity, number of carbon atoms related to alkylation, and sample. These three-way arrays have then be concatenated using low-level data fusion strategy to obtain a new tensor with three new modes: aromaticity, alkylation, and sample. The PARAFAC method has then been applied for the first time to this three-way data structure. A two components decomposition has allowed us to highlight unique samples with unexpected reactivity behaviors throughout hydrotreatment. The obtained loadings led to the identification of the variables responsible for this specific character. This original strategy has provided a fast visualization tool able to highlight simultaneously the impact of the three ionization modes in order to explain the differences between the samples and compare them.

Determination of 10-HDA in royal jelly by ATR-FTMIR and NIR spectral combining with data fusion strategy

The attenuated total reflectance-Fourier transform mid-infrared (ATR-FTMIR) and near-infrared (NIR) spectra of royal jelly samples were collected. Low- and mid-level data fusion strategies in combination with the partial least squares (PLS) regression algorithm were used for quantitative modeling analysis of 10-hydroxy-2-decenoic acid (10-HDA) content in royal jelly samples. In low-level data fusion, each raw spectrum was pre-processed before splicing into a new data matrix for PLS model construction and analysis. In the mid-level data fusion, synergy interval (SI)-PLS was used for variable selection and principal component analysis/independent component analysis was used for feature extraction to obtain variables. The extracted variables were spliced before inputting into the PLS model for modeling and analysis. The results showed that the PLS analysis model constructed by mid-level data fusion is better than the PLS models constructed by independent data and low-level data fusion. Among these models, the PLS model that was constructed by the mid-level data fusion after SI-PLS variable selection had the best 10-HDA content prediction accuracy, with RMSEP = 0.1118(%) and RP = 0.9585. Therefore, a mid-level data fusion strategy based on the ATR-FTMIR and NIR spectra can be used as a reliable tool for 10-HDA quantitation.

Data Fusion of Raman and Near-Infrared Spectroscopies for the Rapid Quantitative Analysis of Methanol Content in Methanol–Gasoline

Rapid analysis of methanol content in methanol–gasoline is of great significance to monitor the methanol–gasoline quality. In this work, two different data-fusion strategies based on Raman and near-infrared (NIR) spectroscopies coupled with partial least square (PLS) were constructed and applied for a rapid and accurate analysis of the methanol content in methanol–gasoline. The Raman and NIR spectra of 49 methanol–gasoline samples were recorded, and the characteristic peaks of the methanol–gasoline samples in Raman and NIR spectroscopies were identified. For spectral data fusion, two different data-fusion strategies based on Raman and NIR spectroscopies coupled with PLS were proposed; normalization was used for low-level data fusion, and variable importance in projection (VIP) was used for mid-level data fusion. The different spectra pretreatment methods, latent variables, and variable importance thresholds of VIP were explored and optimized by 5-fold cross-validation (CV) to optimize the PLS calibration model for methanol content analysis. To further prove the predictive performance and stability of the PLS calibration model based on two data-fusion strategies, four PLS calibration models based on Raman, NIR, and two data-fusion strategies were applied to the quantitative analysis of methanol content in methanol–gasoline. The results show that the predictive performance of PLS calibration models based on the two data-fusion strategies is improved, and the PLS calibration model based on mid-level data fusion strategy gave an excellent predictive performance in methanol content analysis, with coefficients of determination of cross-validation (Rcᵥ²) and validation set (Rᵥ²) of 0.9988 and 0.9905, respectively, and root mean square error of cross-validation (RMSECV) and validation set (RMSEV) of 0.0068 and 0.0288%, respectively. Therefore, data fusion based on Raman and NIR spectroscopies coupled with PLS can give a rapid and accurate quantitative analysis of the methanol content in methanol–gasoline.

Multi-Sensor-Fusion Approach for a Data-Science-Oriented Preventive Health Management System: Concept and Development of a Decentralized Data Collection Approach for Heterogeneous Data Sources.

Investigations in preventive and occupational medicine are often based on the acquisition of data in the customer's daily routine. This requires convenient measurement solutions including physiological, psychological, physical, and sometimes emotional parameters. In this paper, the introduction of a decentralized multi-sensor-fusion approach for a preventive health-management system is described. The aim is the provision of a flexible mobile data-collection platform, which can be used in many different health-care related applications. Different heterogeneous data sources can be integrated and measured data are prepared and transferred to a superordinated data-science-oriented cloud-solution. The presented novel approach focuses on the integration and fusion of different mobile data sources on a mobile data collection system (mDCS). This includes directly coupled wireless sensor devices, indirectly coupled devices offering the datasets via vendor-specific cloud solutions (as e.g., Fitbit, San Francisco, USA and Nokia, Espoo, Finland) and questionnaires to acquire subjective and objective parameters. The mDCS functions as a user-specific interface adapter and data concentrator decentralized from a data-science-oriented processing cloud. A low-level data fusion in the mDCS includes the synchronization of the data sources, the individual selection of required data sets and the execution of pre-processing procedures. Thus, the mDCS increases the availability of the processing cloud and in consequence also of the higher level data-fusion procedures. The developed system can be easily adapted to changing health-care applications by using different sensor combinations. The complex processing for data analysis can be supported and intervention measures can be provided.

A CONCEPT FOR AN AUTOMATED APPROACH OF PUBLIC TRANSPORT VEHICLES TO A BUS STOP

Abstract. This paper discusses the current methods for vehicle self-localization and compares previous findings to the use for urban public traffic vehicles. In specific, requirements for autonomous buses approaching a bus stop are defined. An autonomous system capable of reliable vehicle self-localization running in real-time in a city scenario shall be developed in a future work based on this paper. The comparison of filter-based estimation and graph-based optimization techniques shows that the latter suits the the automated approach to a bus stop in an urban environment the best. Based on these findings, a concept for self-localization of public transport vehicles equipped with a variety of imaging sensors with the help of a digital high definition map is presented. A current method is shown and a concept of improving the localization by inferring semantic information into landmark detection by low-level data fusion is provided. Validation and verification of the proposed fusion approach have to be carried out in the future, but a validation scenario is presented in this work.

Assessing Geographical Origin of Gentiana Rigescens Using Untargeted Chromatographic Fingerprint, Data Fusion and Chemometrics.

Gentiana rigescens Franchet, which is famous for its bitter properties, is a traditional drug of chronic hepatitis and important raw materials for the pharmaceutical industry in China. In the study, high-performance liquid chromatography (HPLC), coupled with diode array detector (DAD) and chemometrics, were used to investigate the chemical geographical variation of G. rigescens and to classify medicinal materials, according to their grown latitudes. The chromatographic fingerprints of 280 individuals and 840 samples from rhizomes, stems, and leaves of four different latitude areas were recorded and analyzed for tracing the geographical origin of medicinal materials. At first, HPLC fingerprints of underground and aerial parts were generated while using reversed-phase liquid chromatography. After the preliminary data exploration, two supervised pattern recognition techniques, random forest (RF) and orthogonal partial least-squares discriminant analysis (OPLS-DA), were applied to the three HPLC fingerprint data sets of rhizomes, stems, and leaves, respectively. Furthermore, fingerprint data sets of aerial and underground parts were separately processed and joined while using two data fusion strategies (“low-level” and “mid-level”). The results showed that classification models that are based OPLS-DA were more efficient than RF models. The classification models using low-level data fusion method built showed considerably good recognition and prediction abilities (the accuracy is higher than 99% and sensibility, specificity, Matthews correlation coefficient, and efficiency range from 0.95 to 1.00). Low-level data fusion strategy combined with OPLS-DA could provide the best discrimination result. In summary, this study explored the latitude variation of phytochemical of G. rigescens and developed a reliable and accurate identification method for G. rigescens that were grown at different latitudes based on untargeted HPLC fingerprint, data fusion, and chemometrics. The study results are meaningful for authentication and the quality control of Chinese medicinal materials.

Data fusion of GC-IMS data and FT-MIR spectra for the authentication of olive oils and honeys-is it worth to go the extra mile?

The potential benefit of data fusion based on different complementary analytical techniques was investigated for two different classification tasks in the field of foodstuff authentication. Sixty-four honey samples from three different botanical origins and 53 extra virgin olive oil samples from three different geographical areas were analyzed by attenuated total reflection IR spectroscopy (ATR/FT-IR) and headspace gas chromatography-ion mobility spectrometry (HS-GC-IMS). The obtained datasets were combined in a low-level data fusion approach with a subsequent multivariate classification by principal component analysis-linear discriminant analysis (PCA-LDA) or partial least squares-discriminant analysis (PLS-DA). Performing a back projection of PCA loadings, the influence of variables in the FT-IR spectra (one-dimensional) and the GC-IMS profiles (two-dimensional) on the discrimination was visualized within the original axis of the two data sources. Validation results of the classification models were compared to the results that could be obtained by using the individual data blocks separately. For both the honey and olive oil samples, a decreased cross-validation error rate and more robust model was obtained due to the low-level data fusion. The results show that data fusion is an effective strategy for improving the classification performance, particularly for challenging classification tasks such as the discrimination of olive oils with different geographical origin. Graphical abstract.

Authentication of the geographical origin of extra-virgin olive oil of the Arbequina cultivar by chromatographic fingerprinting and chemometrics

This paper proposes to use chromatographic fingerprints coupled to multivariate techniques to authenticate the geographical origin of extra-virgin olive oils (EVOO) of the Arbequina botanical variety. This methodology uses the whole or part of the chromatogram as input data for the classification models but does not identify or quantify the chemicals constituents.Arbequina monovarietal EVOOs from three geographical origins were studied: two from adjacent European Protected Designation of Origin areas, Siurana and Les Garrigues, in Catalonia in the northeast of Spain; and the third from the south of Spain (Andalucia and Murcia). Three chromatographic fingerprints of each sample were obtained by both reverse and normal phase liquid chromatography coupled to charged aerosol detector (HPLC-CAD), and high temperature gas chromatography coupled to flame ionization detector [(HT)GC-FID].Principal component analysis (PCA) was used as exploratory technique and soft independent modelling of class analogy (SIMCA) and partial least square-discriminant analysis (PLS-DA) were used as classification methods. High and low-level data fusion strategies were also applied to improve the classification results obtained when the data acquired from each analytical technique were separately used.The results were best for the PLS-DA model with low-level fusion of two techniques (HT)GC-FID with HPLC-CAD, independently of the phase mode. Sensitivity and specificity were 100% in almost all classes, error was 0% for all classes and an inconclusive ratio of just 4% was obtained for the Les Garrigues class due to double assignations.

A holistic approach to authenticate organic sweet oranges (Citrus Sinensis L. cv Osbeck) using different techniques and data fusion

Spain is the second largest European producer of oranges and the fourth in the world. Moreover, in Andalusia, the land dedicated to organic farming of oranges has doubled in the last decade (2006–2016). The great challenge in the organic farming sector is to find a reliable analytical technique (or a set thereof) to verify that the method of production followed the organic production guidelines. A new methodology based on stable isotopes, (poly)phenol profiling and conventional quality data in the edible and non-edible parts of the oranges was developed using data fusion techniques. This system allowed authenticating organic fresh oranges. The application of a low level data fusion and a non-linear algorithm (k-nearest neighbors), obtained 100% specificity and sensitivity in the test set models. This work presented a new focus on dealing with the problem of the authentication of fruits grown under organic farming techniques.

Data fusion strategy in quantitative analysis of spectroscopy relevant to olive oil adulteration

Olive oil adulteration with various less expensive edible oils represents a great danger for consumers. Spectrometry has been used to detect olive oil adulteration with other oil, but we need more robust and accurate model. Therefore, this work investigated the combination of infrared (NIR) and mid infrared (MIR) spectroscopy for the quantification of rapeseed oil in olive oil blends. Furthermore, a partial least squares (PLS) model was established to predict the concentration of the adulterant. Models constructed using baseline correction by combination of standard normal variate (SNV), SG smoothing and vector normalization pretreatments, respectively. Three data fusion strategies (low, mid and high-level) have been applied to take advantage of the synergistic effect of the information obtained from NIR and MIR. We chose algorithm (SPA) to extract spectral features for mid-level data fusion. Binary linear regression used in high-level data fusion. We selected the best pretreatment for final evaluation according to the evaluation parameters (R2 of calibration and validation, RMSECV and RMSEP). NIR, MIR and data fusion models were evaluated by comparing the R2 of validation and RMSEP (root mean square error of prediction). The RMSEP of low-level (3.44), high-level (2.86) data fusion were better than NIR(7.09), MIR(4.04), mid-level(6.09)and the validation coefficient of determination R2 of low-level data fusion (0.975) and high-level data fusion (0.988) are better than the NIR (0.896) and MIR (0.966). Results showed that：(1) NIR and MIR are fast and non-destructive testing tools to detect the extra-virgin olive oil adulteration with rapeseed oil. (2) Low-level data fusion can effectively improve model prediction accuracy. (3) SPA reduced the number of variables, but it did not improved the results. (4) High-level data fusion strategy can be used as a reliable tool for quantitative analysis.

Combining mid infrared spectroscopy and paper spray mass spectrometry in a data fusion model to predict the composition of coffee blends

This paper describes a robust multivariate model for quantifying and characterizing blends of Robusta and Arabica coffees. At different degrees of roasting, 120 ground coffee blends (0.0–33.0%) were formulated. Spectra were obtained by two different techniques, attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy and paper spray mass spectrometry (PS-MS). Partial least squares (PLS) models were built individually with the two types of spectra. Nevertheless, better predictions were obtained by low and medium-level data fusion, taking advantage from the synergy between these two data sets. Data fusion models were improved by variable selection, using genetic algorithms (GA) and ordered predictors selection (OPS). The smallest prediction errors were provided by OPS low-level data fusion model. The number of variables used for regression was reduced from 2145 (full spectra) to 230. Model interpretation was performed by assigning some of the selected variables to specific coffee components, such as trigonelline and chlorogenic acids.

Determination of ethanol content in ethanol-gasoline based on derivative absorption spectrometry and information fusion

The ethanol content in ethanol-gasoline is respectively detected by the first-order derivative UV/vis absorption spectrum, the first-order derivative near infrared (NIR) absorption spectrum and the information fusion method. The backward interval partial least squares (BiPLS) algorithm is used as the feature extraction method, which is established by the partial least squares (PLS) regression model. Based on the information fusion theory, the low level data fusion (LLDF) and mid-level data fusion (MLDF) models are established by the first-order derivative UV/vis and NIR spectra. The analytical results are compared with the related textual references. Thereby, the single-spectral model based on the first-order derivative NIR absorption spectrum has the optimal results, where R p 2 =0.999 1 and RMSEP=0.324 5, while the LLDF after vector normalization (LLDF-VN2) is the optimal multi-spectral fusion model, where R p 2 =0.998 3 and RMSEP=0.498 2. The proposed method can be used to detect the ethanol content in ethanol-gasoline rapidly and provides a better choice for the component detection in mixed oils.

Authentication of Dendrobium Species Using Near-Infrared and Ultraviolet–Visible Spectroscopy with Chemometrics and Data Fusion

ABSTRACTHerbal products produced from multiple plants have special characteristics in the clinical practice of traditional Chinese medicine. These traits provide the opportunity for fraudulent merchants to mix other herbal products similar in appearance into authentic herbal medicine. Shihu is a tonic herbal medicine from the Dendrobium plants with complex botanical origins. In this context, 11 Dendrobium plants including 109 individuals from China were collected for authentication work. Nine species have been described as herbal medicines in the literature while D. hookerianum and D. xichouense are not reported to have medicinal benefits. A key feature of this study was that multiple recognition approaches, based on near-infrared and ultraviolet–visible spectra as well as their combination, were compared to investigate their classification performance. Intuitively, score plots using principal component analysis and hierarchical cluster diagrams were used to evaluate the genetic relationships among these species. Compared with support vector machine discrimination analysis and k-nearest neighbor models, the partial least square discrimination analysis model combined with low-level data fusion provided excellent performance for authentication and was the most robust model with 100% accuracy rates for the training and prediction sets. The results indicated that near-infrared and ultraviolet–visible spectra and their fusion dataset combined with supervised recognition analysis are effective and therefore recommended for the authentication of genuine and sham of herbal Shihu species.

Classification of Paris species according to botanical and geographical origins based on spectroscopic, chromatographic, conventional chemometric analysis and data fusion strategy

Given that the consumption of pure varieties for Paridis Rhizome was increasing and causing a “ripple effect”, i.e. numerous other unofficial Paris species were harvested and traded as the succedaneum of Paridis. The purpose of this study was to establish an effective strategy to distinguish Paris species and trace geographical origins of pure varieties of Paridis samples. A total of 87 batch samples of wild Paris were analyzed by UPLC and FTIR. Among them, 79 batch samples of five species and 40 batch samples of Paris polyphylla Smith var. yunnanensis (Franch.) Hand.-Mazz were used to establish pattern recognition models of PLS-DA, SVM and random forest simultaneously, basing on polyphyllin content, FTIR spectra, UPLC chromatograms, low- and mid-level data fusion strategy. Quantitative analysis results revealed that more obvious fluctuation of polyphyllin content was shown in different species sample than that of pure varieties of Paridis collected from various origins. Pattern recognition models showed that compared with polyphyllin content, FTIR spectra, UPLC chromatograms and low-level data fusion method, mid-level data fusion strategy showed a better accuracy (>92%) in PLS-DA model to classify Paridis samples according to botanical and geographical origins. Similar results were obtained in SVM and random forest models. Thus, the classification of these three pattern recognition models were verified each other and demonstrated that mid-level data fusion strategy combined with chemometrics could recognize different Paris species and trace the pure varieties of Paridis geographical origins, correctly.

Role of Pre-processing in Textual Data Fusion: Learn From the Croydon Tram Tragedy

Tram/train derailment subject to human mistakes makes investments in an advanced control room as well as information gathering system exaggerated. A disaster in Croydon in year 2016 is recent evidence of limitation of the acquired systems to mitigate human shortcoming in disrupted circumstances. One intriguing way of resolution could be is to fuse continuous online textual data obtained from tram travelers and apply the information for early cautioning of risk discovery. This resolution conveys our consideration regarding a resource of data fusion. The focal subject of this paper is to discuss about role of pre-processing ventures in a low-level data fusion that have been distinguished as a pass to avoid time and exertion squandering amid information retrieval. Inclines in online text data pre-processing is reviewed which comes about an outline suggestion that concede traveler's responses through social media channels. The research outcome shows by a case of data fusion could go about as an impetus to railway industry to effectively partake in data exploration and information investigation.

Near-infrared spectroscopy and X-ray fluorescence data fusion for olive leaf analysis and crop nutritional status determination

Leaf analysis is a useful way of diagnosing the nutritional status of the plants and therefore fast methods of analysis are demanded to aid in fertilization management decisions. In this work, a strategy based on the combined use of near-infrared spectroscopy (NIR) and portable energy dispersive X-Ray Fluorescence (EDXRF) is proposed as a suitable cheap and rapid alternative to traditional wet analytical methodologies. The approach has the major benefit of minimal sample preparation since leaves need to be only dried and ground. The ability of both techniques individually and applying two strategies of data fusion for the prediction of the most important plant nutrients, namely N, P, K, Ca, Mg, Mn, Zn, and B was tested. Predictive models were constructed using Partial Least Squares (PLS) to correlate the spectra with the nutrient contents. Models of unequal prediction performance in terms of the ratio of predictive deviation (RPD) were obtained for the different parameters when considering both techniques separately. Low-level data fusion, which consists of a concatenation of the raw data from both techniques, showed little improvement and even decreased the predictive ability for some elements. Better results were obtained with mid-level data fusion, that is, merging data after a feature extraction step performed by means of Principal components analysis (PCA). The results show that a fair quantitative prediction is possible for Ca, K and Mn with RPDs ≥ 2 for external validation, whereas models for N and P allowed a semiquantitative estimation. Mg and B models were less satisfactory and can be used only for distinguish between low and high levels, while Zn content cannot be predicted. Finally, the potential of the fusion of FT-NIR and EDXRF spectroscopic data for the fast screening of olive crop nutritional status has been tested. Deficiencies in important elements like N and K has been successfully detected.

Chemical profiling and multivariate data fusion methods for the identification of the botanical origin of honey

The characterization of 72 Italian honey samples from 8 botanical varieties was carried out by a comprehensive approach exploiting data fusion of IR, NIR and Raman spectroscopies, Proton Transfer Reaction – Time of Flight – Mass Spectrometry (PTR-MS) and electronic nose. High-, mid- and low-level data fusion approaches were tested to verify if the combination of several analytical sources can improve the classification ability of honeys from different botanical origins. Classification was performed on the fused data by Partial Least Squares – Discriminant Analysis; a strict validation protocol was used to estimate the predictive performances of the models. The best results were obtained with high-level data fusion combining Raman and NIR spectroscopy and PTR-MS, with classification performances better than those obtained on single analytical sources (accuracy of 99% and 100% on test and training samples respectively). The combination of just three analytical sources assures a limited time of analysis.

Exploiting the synergistic effect of concurrent data signals: Low-level fusion of liquid chromatographic with dual detection data

A low-level data fusion strategy was developed and implemented for data processing of second-order liquid chromatographic data with dual detection, i.e. absorbance and fluorescence monitoring. The synergistic effect of coupling individual information provided by two different detectors was evaluated by analyzing the results gathered after the application of a series of data preprocessing steps and chemometric resolution. The chemometric modeling involved data analysis by MCR-ALS, PARAFAC and N-PLS. Their ability to handle the new data block was assessed through the estimation of the analytical figures of merits achieved in the prediction of a validation set containing fifteen fluorescent and non-fluorescent veterinary active ingredients that can be found in poultry litter. Eventually, the feasibility of the application of the fusion strategy to real poultry litter samples containing the studied compounds was verified.

FT-MIR and UV–vis data fusion strategy for origins discrimination of wild Paris Polyphylla Smith var. yunnanensis

Paris Polyphylla Smith var. yunnanensis (Franch.) Hand.-Mazz has multiple therapeutic properties and the origins may affect clinical efficacy. Tracing the geographical origin is important to the authentication and quality assessment of this species. 177 wild samples collected from central, southeast and northwest Yunnan Province, China, were analyzed by single analytical method and data fusion strategies (low- and mid-levels) using Fourier transform mid-infrared (FT-MIR) and ultraviolet-visible (UV–vis) spectroscopies combined with chemometrics (partial least squares discrimination analysis (PLS-DA) and support vector machines grid search (SVM-GS)), for categorizing samples from different geographic origins. According to the results, mid-level data fusion strategy presented a better generalization performance and accuracy rates based on latent variables selected by PLS-DA than single analytical method and low-level data fusion strategy. Accuracy rates were almost 100% when both of the PLS-DA and SVM-GS were employed for classifying samples picked from southeast and northwest districts based on mid-level dataset. For samples collected from central of Yunnan where was divided into seven categories in this paper, the accuracy rates of training set and test set of PLS-DA and SVM-GS were preferable (>87%). Based on the mid-level data set, both of the classification results of PLS-DA and SVM-GS presented satisfying accuracy for 177 samples. Additionally, as small as possible parameters showed in mid-level data set, it suggested that this method was robust and generalized. Therefore, the comprehensive method was established for the origin traceability of wild P. Polyphylla Smith var. yunnanensis, which is meaningful for the quality control of herbal medicines.