Partial Data Sets Research Articles

Disambiguation-based partial label feature selection via feature dependency and label consistency

Partial label learning refers to the issue that each training sample corresponds to a candidate label set containing only one valid label. Feature selection can be viewed as an effective pre-processing technology to improve the generalization performance of learning models, while the partial label feature selection task is challenging due to ambiguous labeling information. To this end, this paper utilizes granular ball computing and neighborhood rough sets to put forward a disambiguation-based partial label feature selection algorithm via feature dependency and label consistency. Firstly, the proposed algorithm performs an adaptive neighbor aggregation based on granular balls to disambiguate candidate labels. Adaptive neighbors are flexible to make more efficient learning performance. Then, considering the labeling confidence induced by disambiguation, the significance of each feature is evaluated by fusing the feature dependency from the neighborhood granularity and the label consistency among the nearest neighbors. Neighborhood rough sets directly handle continuous features that can reduce the adverse impact of data discretization on feature selection. The label consistency among adjacent samples is important for measuring feature discrimination. The fusion of feature dependency and label consistency can facilitate performance improvement. Finally, experiments conducted on eight controlled UCI and five real-world partial label datasets demonstrate that the proposed algorithm can improve the generalization performance of partial label learning and achieve superior performance compared to the partial label feature selection methods.

A Neural Network-Based Partial Fingerprint Image Identification Method for Crime Scenes

Fingerprints are the most widely used of all biological characteristics in public safety and forensic identification. However, fingerprint images extracted from the crime scene are incomplete. On the one hand, due to the lack of effective area in partial fingerprint images, the extracted features are insufficient. On the other hand, a broken ridge may lead to a large number of false feature points, which affect the accuracy of fingerprint recognition. Existing fingerprint identification methods are not ideal for partial fingerprint identification. To overcome these problems, this paper proposes an attention-based partial fingerprint identification model named APFI. Firstly, the algorithm utilizes the residual network (ResNet) for feature descriptor extraction, which generates a representation of spatial information on fingerprint expression. Secondly, the channel attention module is inserted into the proposed model to obtain more accurate fingerprint feature information from the residual block. Then, to improve the identification accuracy of partial fingerprints, the angular distance between features is used to calculate the similarity of fingerprints. Finally, the proposed model is trained and validated on a home-made partial fingerprint image dataset. Experiments on the home-made fingerprint datasets and the NIST-SD4 datasets show that the partial fingerprint identification method proposed in this paper has higher identification accuracy than other state-of-the-art methods.

Occluded Person Re-Identification With Pose Estimation Correction and Feature Reconstruction

In the real-world surveillance system application, the accuracy of person re-identification (Re-ID) still suffers from occlusion. Occluded Re-ID aims to retrieve occluded person images from multiple cameras that do not overlap. To address this issue, we present a multi-branch feature enhancement network in this paper. It has the pose correction module (PCM), the feature reconstruction module (FRM), and the part align module (PAM). PCM builds the key-points confidence corrected mechanism of the self-adaptive weights learning before the part matching stage to strengthen part-level features in the non-occluded region and weaken ones in the occluded region. FRM reconstructs the feature distribution of inter-domain and class-domain using the maximum mean discrepancy to generalize the global features. The proposed feature separation mechanism enables the two branches to attend to distinct pedestrian feature. To increase the power of representation, PAM combines part-level and global features in an outer product. The experimental results on three holistic datasets, two partial datasets, and two occluded datasets demonstrate that our method is superior. It significantly outperforms state-of-the-art by 5.7% and 4.4% rank-1 scores on Occluded-Duke and Occluded-REID, respectively.

Improving Generalized Zero-Shot Learning SSVEP Classification Performance From Data-Efficient Perspective.

Generalized zero-shot learning (GZSL) has significantly reduced the training requirements for steady-state visual evoked potential (SSVEP) based brain-computer interfaces (BCIs). Traditional methods require complete class data sets for training, but GZSL allows for only partial class data sets, dividing them into 'seen' (those with training data) and 'unseen' classes (those without training data). However, inefficient utilization of SSVEP data limits the accuracy and information transfer rate (ITR) of existing GZSL methods. To this end, we proposed a framework for more effective utilization of SSVEP data at three systematically combined levels: data acquisition, feature extraction, and decision-making. First, prevalent SSVEP-based BCIs overlook the inter-subject variance in visual latency and employ fixed sampling starting time (SST). We introduced a dynamic sampling starting time (DSST) strategy at the data acquisition level. This strategy uses the classification results on the validation set to find the optimal sampling starting time (OSST) for each subject. In addition, we developed a Transformer structure to capture the global information of input data for compensating the small receptive field of existing networks. The global receptive fields of the Transformer can adequately process the information from longer input sequences. For the decision-making level, we designed a classifier selection strategy that can automatically select the optimal classifier for the seen and unseen classes, respectively. We also proposed a training procedure to make the above solutions in conjunction with each other. Our method was validated on three public datasets and outperformed the state-of-the-art (SOTA) methods. Crucially, we also outperformed the representative methods that require training data for all classes.

Occluded Person Re-Identification via Defending Against Attacks From Obstacles

Due to incomplete appearance features, the identity matching of occluded pedestrians under multiple cross-camera views is a long-term challenge. Although existing re-identification (re-ID) solutions of occluded pedestrians have made significant progress, most of them achieve accurate identity matching by extracting pedestrian appearance features from unoccluded areas. However, when a pedestrian is partially blocked by the body of another pedestrian, existing methods cannot accurately determine whether the unoccluded body parts belong to the target pedestrian, which brings great difficulties to pedestrian identity matching. To alleviate this problem, this paper introduces the idea of adversarial attack into occluded person re-ID and proposes an adversarial training framework that can defend against attacks from obstacles to resist the interference of obstacles on pedestrian identity matching. Unlike existing solutions, the proposed framework is not limited to extracting features of unoccluded human body areas to achieve occluded person re-ID, but explores how to make the re-ID model more resistant to obstacles. In the proposed framework, the occluded pedestrian images are regarded as adversarial examples and used to attack model training. If the trained model can defend against this kind of attack, its generalization is significantly improved, and the above-mentioned issues are also effectively solved. Specifically, a single-branch dual-stream collaborative network is designed. With the cooperation of the pre-trained verification guidance network, the model realizes the attack and defense of adversarial samples. This work broadens research horizons in robust model design of occluded person re-ID, and expands the scope of adversarial attacks. Compared with existing solutions, a lot of experimental results confirm that the proposed solution achieves better performance on two occluded re-ID datasets and two partial re-ID datasets.

Hybrid optimization with unconstrained variables on partial point cloud registration

3D point cloud registration is a fundamental problem in computer vision (CV) and computer graphics (CG). Recently, a series of learning-based algorithms have been proposed to show the advantages in registration accuracy and inference speed. However, those learning-based methods usually ignore transformations with constrained rotations and translations in registration. In this paper, we propose a novel hybrid optimization method to solve the constrained rotational and translational transformations. A mapping function is introduced to deal with the restrained variables in optimization. Our method achieves superior performance on the Multi-View Partial Point dataset, which won the first place on the registration challenge in ICCV 2021. The method is also validated on the synthetic datasets ModelNet, ICL-NUIM, and the realistic 3DMatch dataset. We demonstrate that the global optimization methods still have great potential research for point cloud registration. The code is available at https://github.com/Dizzy-cell/HOUV.

Deriving parametric and probabilistic Kd values for fluoroquinolones in soils

The evaluation of the sorption affinity of fluoroquinolone antibiotics (FQs) in soils, by means of the derivation of solid-liquid distribution coefficients (Kd), is a valuable information for assessing their environmental mobility. Aiming to develop Kd (FQ) prediction tools in soils, in the first stage of this study we constructed a Kd (FQ) sorption dataset using current literature data. Furthermore, additional sorption and desorption data for norfloxacin were obtained in seven different soils of contrasting properties. Sorption isotherms of norfloxacin were linear under the experimental conditions tested and desorption percentages increased for scenarios in which low sorption was noted. Sorption tests in the same soils were then extended to ciprofloxacin, enrofloxacin and ofloxacin and pooled in the dataset, revealing comparable Kd (FQ) values among the FQ tested after analyzing the overall dataset consisting in 312 entries of Kd (FQ). A partial least square (PLS) regression model was then developed to predict values of Kd (FQ) based on specific relevant soil properties (i.e., pH, cation exchange capacity and organic carbon and texture information), and, for the first time, FQ properties (fraction of cationic FQ species) affecting sorption. Additionally, probabilistic, Kd (FQ) best estimates in soils were derived through cumulative distribution functions (CDFs) for the overall and for partial datasets created by grouping Kd (FQ) values according to key soil properties affecting FQ sorption (i.e., pH, organic carbon content and texture information). This latter approach permitted to derive more representative Kd (FQ) best estimates for the soils to be assessed, and with a lower related variability than that derived from the overall dataset. Best estimates Kd (FQ) values were > 1000 L kg−1 for most acidic to neutral soils, suggesting strong sorption, although lower sorption and thus higher environmental mobility may be expected in scenarios with soils with alkaline pH, low OC and high sand contents. SynopsisThis study aims to derive parametric and probabilistic Kd values for fluoroquinolone antibiotics in soils on the basis of a few relevant soil physicochemical properties.

Top-k Partial Label Machine.

To deal with ambiguities in partial label learning (PLL), the existing PLL methods implement disambiguations, by either identifying the ground-truth label or averaging the candidate labels. However, these methods can be easily misled by the false-positive labels in the candidate label set. We find that these ambiguities often originate from the noise caused by highly correlated or overlapping candidate labels, which leads to the difficulty in identifying the ground-truth label on the first attempt. To give the trained models more tolerance, we first propose the top-k partial loss and convex top-k partial hinge loss. Based on the losses, we present a novel top-k partial label machine (TPLM) for partial label classification. An efficient optimization algorithm is proposed based on accelerated proximal stochastic dual coordinate ascent (Prox-SDCA) and linear programming (LP). Moreover, we present a theoretical analysis of the generalization error for TPLM. Comprehensive experiments on both controlled UCI datasets and real-world partial label datasets demonstrate that the proposed method is superior to the state-of-the-art approaches.

Retrospective Fat Suppression for Lung Radiotherapy Planning with CycleGANs

<h3>Purpose/Objective(s)</h3> MR-based radiotherapy planning and MR-guided radiotherapy are gaining traction. In current practice, several MR sequences are used for planning, including T2 weighted nonfat-suppressed (NFS) images that show fat between organs at risk and fat-suppressed (FS) images that optimize tumor conspicuity. To spare time and avoid potential motion between sequences, retrospective fat suppression with AI has been explored. Rowland et al. found that models learned to introduce artifacts present in training data. In this study we investigate whether training models on a smaller dataset without artifacts produces better results than training on a large dataset including artifacts. <h3>Materials/Methods</h3> 21 lung cancer patients were each scanned twice on a 1.5T Aera. At each session, 2D TSE, navigator triggered, T2-weighted images (TE/TR=102/5200ms, 30 × 3.5mm slices with matrix 320 × 320, 1.25 × 1.25mm pixels, TA ∼6 minutes) were acquired both with and without fat suppression. Ten NFS/FS slice pairs were randomly selected as the test dataset and remaining images were used for training. Three CycleGANs with identical structures were trained as follows. Full model was trained on the full dataset (910 image pairs) for 400 epochs. For the partial model images with fat suppression artifacts were removed (resulting in 500 pairs) and the model was trained for 720 epochs. Transfer model was trained on full dataset for 200 epochs, then partial dataset for 360 epochs. This selection results in equal training time for all models. Fat and tumor intensities are similar, thus fat suppression can cause undesirable reduction in tumor signal. To assess changes in both areas, tumor and subcutaneous fat were delineated on test images and mean intensity change compared to NFS image in delineated areas was calculated. Also, an MR radiographer and a medical physicist scored quality and fidelity of the tumor region in 10 sets of generated FS images against a corresponding image with prospective FS on a 5-point Likert scale. <h3>Results</h3> Table 1 shows results of qualitative and quantitative analyses. The partial and transfer models were better than full network at suppressing fat without introducing artifacts. Within tumors partial model provided superior sharpness, while transfer model better maintained intensity level and fidelity. <h3>Conclusion</h3> Despite differences being small, transfer learning approach produced the result closest to the real FS, indicating that both quality and quantity of training data are important.

Advances in non-invasive biosensing measures to monitor wound healing progression

Impaired wound healing is a significant financial and medical burden. The synthesis and deposition of extracellular matrix (ECM) in a new wound is a dynamic process that is constantly changing and adapting to the biochemical and biomechanical signaling from the extracellular microenvironments of the wound. This drives either a regenerative or fibrotic and scar-forming healing outcome. Disruptions in ECM deposition, structure, and composition lead to impaired healing in diseased states, such as in diabetes. Valid measures of the principal determinants of successful ECM deposition and wound healing include lack of bacterial contamination, good tissue perfusion, and reduced mechanical injury and strain. These measures are used by wound-care providers to intervene upon the healing wound to steer healing toward a more functional phenotype with improved structural integrity and healing outcomes and to prevent adverse wound developments. In this review, we discuss bioengineering advances in 1) non-invasive detection of biologic and physiologic factors of the healing wound, 2) visualizing and modeling the ECM, and 3) computational tools that efficiently evaluate the complex data acquired from the wounds based on basic science, preclinical, translational and clinical studies, that would allow us to prognosticate healing outcomes and intervene effectively. We focus on bioelectronics and biologic interfaces of the sensors and actuators for real time biosensing and actuation of the tissues. We also discuss high-resolution, advanced imaging techniques, which go beyond traditional confocal and fluorescence microscopy to visualize microscopic details of the composition of the wound matrix, linearity of collagen, and live tracking of components within the wound microenvironment. Computational modeling of the wound matrix, including partial differential equation datasets as well as machine learning models that can serve as powerful tools for physicians to guide their decision-making process are discussed.

Suvorexant maintenance enhances the reinforcing but not subjective and physiological effects of intravenous cocaine in humans

Preclinical research has sought to understand the role of the orexin system in cocaine addiction given the connection between orexin producing cells in the lateral hypothalamus and brain limbic areas. Exogenous administration of orexin peptides increased cocaine self-administration whereas selective orexin-1 receptor antagonists reduced cocaine self-administration in non-human animals. The first clinically available orexin antagonist, suvorexant (a dual orexin-1 and orexin-2 receptor antagonist), attenuated motivation for cocaine and cocaine conditioned place preference, as well as cocaine-associated impulsive responding, in rodents. This study aimed to translate those preclinical findings and determine whether suvorexant maintenance altered the pharmacodynamic effects of cocaine in humans. Seven non-treatment seeking subjects with cocaine use disorder completed this within-subject human laboratory study, and a partial data set was obtained from one additional subject. Subjects were maintained for at least three days on 0, 5, 10 and 20 mg oral suvorexant administered at 2230 h daily in random order. Subjects completed experimental sessions in which cocaine self-administration of 0, 10 and 30 mg/70 kg of intravenous cocaine was evaluated on a concurrent progressive ratio drug versus money choice task. Subjective and physiological effects of cocaine were also determined. Cocaine functioned as a reinforcer and produced prototypic dose-related subjective and physiological effects (e.g., increased ratings of “Stimulated” and heart rate). Suvorexant (10, 20 mg) increased self-administration of 10 mg/70 kg cocaine and decreased oral temperature but did not significantly alter any other effects of cocaine. Future research may seek to evaluate the effects of orexin-1 selective antagonists in combination with cocaine.

Identifying Visible Parts via Pose Estimation for Occluded Person Re-Identification.

We focus on the occlusion problem in person re-identification (re-id), which is one of the main challenges in real-world person retrieval scenarios. Previous methods on the occluded re-id problem usually assume that only the probes are occluded, thereby removing occlusions by manually cropping. However, this may not always hold in practice. This article relaxes this assumption and investigates a more general occlusion problem, where both the probe and gallery images could be occluded. The key to this challenging problem is depressing the noise information by identifying bodies and occlusions. We propose to incorporate the pose information into the re-id framework, which benefits the model in three aspects. First, it provides the location of the body. We then design a Pose-Masked Feature Branch to make our model focus on the body region only and filter those noise features brought by occlusions. Second, the estimated pose reveals which body parts are visible, giving us a hint to construct more informative person features. We propose a Pose-Embedded Feature Branch to adaptively re-calibrate channel-wise feature responses based on the visible body parts. Third, in testing, the estimated pose indicates which regions are informative and reliable for both probe and gallery images. Then we explicitly split the extracted spatial feature into parts. Only part features from those commonly visible parts are utilized in the retrieval. To better evaluate the performances of the occluded re-id, we also propose a large-scale data set for the occluded re-id with more than 35 000 images, namely Occluded-DukeMTMC. Extensive experiments show our approach surpasses previous methods on the occluded, partial, and non-occluded re-id data sets.

A Lightweight Subgraph-Based Deep Learning Approach for Fall Recognition.

Falls pose a great danger to social development, especially to the elderly population. When a fall occurs, the body’s center of gravity moves from a high position to a low position, and the magnitude of change varies among body parts. Most existing fall recognition methods based on deep learning have not yet considered the differences between the movement and the change in amplitude of each body part. Besides, some problems exist such as complicated design, slow detection speed, and lack of timeliness. To alleviate these problems, a lightweight subgraph-based deep learning method utilizing skeleton information for fall recognition is proposed in this paper. The skeleton information of the human body is extracted by OpenPose, and an end-to-end lightweight subgraph-based network is designed. Sub-graph division and sub-graph attention modules are introduced to add a larger perceptual field while maintaining its lightweight characteristics. A multi-scale temporal convolution module is also designed to extract and fuse multi-scale temporal features, which enriches the feature representation. The proposed method is evaluated on a partial fall dataset collected in NTU and on two public datasets, and outperforms existing methods. It indicates that the proposed method is accurate and lightweight, which means it is suitable for real-time detection and rapid response to falls.

Privacy-Preserving and Efficient Verification of the Outcome in Genome-Wide Association Studies.

Providing provenance in scientific workflows is essential for reproducibility and auditability purposes. In this work, we propose a framework that verifies the correctness of the aggregate statistics obtained as a result of a genome-wide association study (GWAS) conducted by a researcher while protecting individuals' privacy in the researcher's dataset. In GWAS, the goal of the researcher is to identify highly associated point mutations (variants) with a given phenotype. The researcher publishes the workflow of the conducted study, its output, and associated metadata. They keep the research dataset private while providing, as part of the metadata, a partial noisy dataset (that achieves local differential privacy). To check the correctness of the workflow output, a verifier makes use of the workflow, its metadata, and results of another GWAS (conducted using publicly available datasets) to distinguish between correct statistics and incorrect ones. For evaluation, we use real genomic data and show that the correctness of the workflow output can be verified with high accuracy even when the aggregate statistics of a small number of variants are provided. We also quantify the privacy leakage due to the provided workflow and its associated metadata and show that the additional privacy risk due to the provided metadata does not increase the existing privacy risk due to sharing of the research results. Thus, our results show that the workflow output (i.e., research results) can be verified with high confidence in a privacy-preserving way. We believe that this work will be a valuable step towards providing provenance in a privacy-preserving way while providing guarantees to the users about the correctness of the results.

Examining the Levels of Violence in Mental Health Trusts

AimsA recent NICE report stated that there were 68,683 assaults reported by NHS staff between 2013 and 2014. 69% of these were in the mental health or learning disability setting. We sought to explore the number of violent incidents within mental health trusts across England and to understand whether the levels of violence against staff have increased, decreased, or remained the same between the years 2014 to 2019. We also looked at whether a change in bed numbers correlated with the levels of violence experienced.MethodsMental Health Trusts in England were identified, and Freedom of Information requests were sent to them. We asked for the numbers of sexual and physical violence between the years 2014 and 2020, broken down by outpatient and inpatient setting. Using bed data from NHS England we looked at whether there was a correlation with violence.ResultsOut of the 53 trusts we approached with freedom of information requests, 43 returned responses with data that could be used for analysis. Data sets were often incomplete, especially for the earlier years requested. The total number of violent incidents from the 43 trusts was 24,393, in the year 2014. There was an increase to 37,907 by the year 2019, which may, in part, be explained by more complete data. Over the same time period, there was a decrease in bed numbers. Average number of episodes of violence per bed increased over 2014 to 2019 from 2 to 2.5, but the increase was not statistically significant. From our data, a correlation between the decrease in bed numbers and increase in rates of violence cannot be drawn.ConclusionThe high number of violent incidents within the mental health setting remain troubling, particularly when taking into account that this analysis represented only a partial data set. This limitation, together other data robustness issues, including the probability of under reporting by staff mean that firm conclusions cannot be drawn. This remains an area where urgent further research is needed, both to identify the extent of the problem, and to probe the impact violence has on staff and patients.

IRDA: Incremental Reinforcement Learning for Dynamic Resource Allocation

Resource allocation problems often manifest as online decision-making tasks where the proper allocation strategy depends on the understanding of the allocation environment and resource workload. Most existing resource allocation methods are based on meticulously designed heuristics which ignore the patterns of incoming tasks, so the dynamics of incoming tasks cannot be properly handled. To address this problem, we mine the task patterns from the large volume of historical allocation data and propose a reinforcement learning model termed IRDA to learn the allocation strategy in an incremental way. We observe that historical allocation data is usually generated from the daily repeated operations, which is not independent and identically distributed. Training with the partial dataset can make the strategy converged already, thereby wasting the remaining data. To improve the learning efficiency, we partition the whole historical big dataset into multi-batch datasets, which forces the agent continuously to ‘explore’ and learn on the distinct state spaces. We apply the proposed method to handle baggage carousel allocation at Hong Kong International Airport (HKIA). The experimental results show that IRDA improves the baggage carousel resource utilization by around 51.86% compared to the current baggage carousel allocation system at HKIA.

Structural damage detection via phase space based manifold learning under changing environmental and operational conditions

The feasibility and performance of existing vibration-based damage detection methods to real world civil engineering structures are inevitably affected by the varying environmental and operational conditions. Reliable damage detection methods with damage features that are sensitive to structural condition change but robust to environmental and loading effects are desirable for practical applications. This paper proposes a novel structural damage detection approach based on manifold learning for the effective condition assessment of real-world structures under environmental and operational conditions. The phase space representation of the vibration characteristics is reconstructed using the identified natural frequencies of structures. Then, the intrinsic nonlinear manifold between the environmental variables and natural frequencies in the high dimensional phase space is projected to a low-dimensional representation via manifold learning. The Gaussian process regression technique is introduced to extract reliable damage index from the learned manifold structure. The effectiveness and superiority of the proposed approach are demonstrated by two real-world engineering structures, that is, the Dowling Hall Footbridge and Z24 bridge. Damage detection results obtained from the proposed approach are compared with those from the current state-of-the-art Kernel PCA method, which is a representative nonlinear dimensionality reduction method to alleviate the environmental effects. The results demonstrate that the proposed approach is sensitive to structural damage but insensitive to changes in environmental and operational conditions. More importantly, the nonlinear environmental effects can be efficiently characterized by the proposed approach, using only partial datasets with environmental variations in the training datasets.

Single-support serial isomorphous replacement phasing.

The use of single isomorphous replacement (SIR) has become less widespread due to difficulties in sample preparation and the identification of isomorphous native and derivative data sets. Non-isomorphism becomes even more problematic in serial experiments, because it adds natural inter-crystal non-isomorphism to heavy-atom-soaking-induced non-isomorphism. Here, a method that can successfully address these issues (and indeed can benefit from differences in heavy-atom occupancy) and additionally significantly simplifies the SIR experiment is presented. A single heavy-atom soak into a microcrystalline slurry is performed, followed by automated serial data collection of partial data sets. This produces a set of data collections with a gradient of heavy-atom occupancies, which are reflected in differential merging statistics. These differences can be exploited by an optimized genetic algorithm to segregate the pool of data sets into `native' and `derivative' groups, which can then be used to successfully determine phases experimentally by SIR.

Considering similarity and the rating conversion of neighbors on neural collaborative filtering.

One of the most popular recommender system techniques is collaborative filtering (CF). Nowadays, many researchers apply a neural network with CF, but few focus on the neighbors’ concept of CF. This work needs to consider two major issues: the similarity levels between the neighbors and the target user and the user’s rating pattern conversion. Because different neighbors have a different influence on the target user and different users usually have a different rating pattern, the ratings directly utilized by the neighbor’s preference pattern may be incorrect for the target user. Under two main issues, we try to accomplish three main ideas of CF’s prediction: the similarity between users’ levels, the neighbor’s rating, and the rating conversion. Thus, we propose three main modules, the rating conversion module, the similarity module, and the prediction module, to solve the two issues mentioned above. The rating conversion module projects the neighbor’s rating into the target user’s aspect. The similarity module uses the users’ attentions to compute similarity levels between users. Finally, these similarity levels and the converted ratings are integrated to perform the prediction. The proposed method is compared with the current CF with friends and latent factor model using two types of datasets: real-world and synthetic datasets. We evaluate N neighbors and all neighbors on real-world datasets to prove the number of neighbor is important. Moreover, the performance of the rating conversion module is also evaluated. The proposed method simulates the full rating datasets and the partial rating dataset to compare the effectiveness of using different types of distribution and dataset size. The experimental results demonstrate that the proposed method effectively outperformed the baselines using ranking evaluation and prediction accuracy on real-world and synthetic datasets. Besides, The effectiveness of using different the number of neighbors depends on the quality of neighbors.

Hyperspectral imaging facilitates early detection of Orobanche cumana below-ground parasitism on sunflower under field conditions

• Early detection of O. cumana in soil subsurface facilitates site-specific control. • Hyperspectral imaging of host leaves, in field conditions, detects O. cumana at an early parasitism stage. • Leaf segmentation improves classification accuracy. • Multispectral data is sufficient for classification of parasitized plants. Sunflower broomrape ( Orobanche cumana ) is a root parasitic weed that severely limits sunflower yield in large areas of Europe and Asia. Early detection of the parasite can facilitate site-specific control of this weed. However, most of its life-cycle takes place in the soil sub-surface and by the time that O . cumana shoots emerge, the damage to the crop is irreversible. The main aim of this study was to evaluate the potential use of hyperspectral imaging for the early detection of parasitism by monitoring changes in spectra obtained from the host plants. A field experiment was conducted on infested and non-infested sunflower plants, imaged by a ground-based hyperspectral camera at two early parasitism stages that are relevant for herbicide application. A logistic regression model was used to classify infected and non-infected plants, 31 and 38 days after sunflower planting, with 76 and 89% accuracy, respectively. A partial dataset, containing only 10 spectral bands of the hyperspectral dataset, gave 69 and 82% accuracy, indicating the potential of multi-spectral sensors for the detection task. Sampling pixels from specific sunflower leaf segments improved the classification compared to non-specific sampling. This study thus contributes to establishing a basis for future development of site-specific weed management of O. cumana and of other broomrape species.