Cluster Compactness Research Articles

Fractional Derivative to Symmetrically Extend the Memory of Fuzzy C-Means

The fuzzy C-means (FCM) clustering algorithm is a widely used unsupervised learning method known for its ability to identify natural groupings within datasets. While effective in many cases, FCM faces challenges such as sensitivity to initial cluster assignments, slow convergence, and difficulty in handling non-linear and overlapping clusters. Aimed at these limitations, this paper introduces a novel fractional fuzzy C-means (Frac-FCM) algorithm, which incorporates fractional derivatives into the FCM framework. By capturing non-local dependencies and long memory effects, fractional derivatives offer a more flexible and precise representation of data relationships, making the method more suitable for complex datasets. Additionally, a genetic algorithm (GA) is employed to optimize a new least-squares objective function that emphasizes the geometric properties of clusters, particularly focusing on the Fukuyama–Sugeno and Xie–Beni indices, thereby enhancing the balance between cluster compactness and separation. Furthermore, the Frac-FCM algorithm is evaluated on several benchmark datasets, including Iris, Seed, and Statlog, and compared against traditional methods like K-means, SOM, GMM, and FCM. The results indicate that Frac-FCM consistently outperforms these methods in terms of the Silhouette and Dunn indices. For instance, Frac-FCM achieves higher Silhouette scores of most cases, indicating more distinct and well-separated clusters. Dunn’s index further shows that Frac-FCM generates clusters that are better separated, surpassing the performance of traditional methods. These findings highlight the robustness and superior clustering performance of Frac-FCM. The Friedman test was employed to enhance and validate the effectiveness of Frac-FCM.

Mechanical Leaf Removal for Improved Botrytis Bunch Rot Control in Vitis vinifera 'Pinot gris' and 'Pinot noir' Grapevines in the Northeastern United States.

Late-season bunch rot can cause substantial yield loss in grapevines grown in humid regions. Fruit-zone leaf removal has been widely used to reduce bunch rot and pesticide applications through improvements in canopy microclimate and grape cluster morphology. In this study, we evaluated whether mechanical leaf removal can be a valid alternative to a labor-intensive manual application by comparing prebloom manual (PB-MA) and mechanical (PB-ME) leaf removal. We also evaluated the effects of the timing of mechanical application, prebloom (PB-ME) versus fruit set (FS-ME), on fruit traits and bunch rot, caused by Botrytis cinerea. Our trials were conducted on two Vitis vinifera 'Pinot noir' and 'Pinot gris' vineyards in the Northeastern United States over two seasons (2017 to 2018). Major findings were overall consistent between cultivars and years. Leaf removal provided reductions in fruit-zone canopy density regardless of method or timing. In general, PB-ME provided similar shifts in cluster morphological traits to PB-MA, including lower number of berries per cluster, cluster compactness, and cluster weight compared with control (no leaf removal) vines. At harvest, both prebloom leaf removal methods equally reduced Botrytis bunch rot severity, whereas Botrytis bunch rot incidence in Pinot noir was lowest for PB-ME in 1 year and PB-MA in the next year. When comparing timing of mechanical leaf removal, FS-ME provided Botrytis bunch rot reductions similar to PB-ME, without effects on cluster weight. Thus, under our growing conditions, FS-ME was considered the best mechanical leaf removal option to help manage Botrytis bunch rot without causing undesirable yield reductions.

Differentiation of long Non-Coding RNA expression profiles in three Fruiting stages of grape

Grapes are considered a crucial fruit crop with extensive uses globally. Cluster compactness is an undesirable trait for the productivity of Yaghooti grape, and it reduces its desirability among consumers. The RNA-Seq data were analyzed in three stages of cluster development using the FEELnc software, leading to the identification of 849 lncRNAs. 183 lncRNAs were differentially expressed during cluster development stages. The GO and KEGG enrichment analyses of these lncRNAs revealed that they target 1,814 genes, including CKX, PG, PME, NPC2, and UGT. The analysis demonstrated a relationship between these lncRNAs and key processes such as grape growth and development, secondary metabolite synthesis, and resistance to both biotic and abiotic stresses. These findings, combined with molecular experiments on lncRNA interactions with other regulatory factors, could enhance Yaghooti grape quality and decrease cluster compactness.

Numerical investigation of the impact of the diameter and rotational motion of a multi-blade rotor on the hydrokinetic combined turbine performance

The vertical axis combined hydrokinetic turbine (VACHT) presents promising applications in marine environments due to its high self-starting performance and efficiency. However, traditional VACHTs suffer from significant performance losses in the outer rotor when the diameter ratio is large and the startup dead zone issue of the inner Savonius rotor. To address these challenges, this study proposes a VACHT composed of an H-type rotor and a multi-blade (M-type) rotor, categorized into two structures based on counter-rotating and co-rotating operations. Numerical simulations reveal that the diameter ratio (f) has a greater impact on VACHT performance than the tip speed ratio. Counter-rotating allows the VACHTs to accommodate larger diameter ratios (f > 0.6) than co-rotation. The diameter ratio primarily influences the interference between the inner and outer rotors. The M-type rotor enhances the self-starting performance of both VACHTs, providing enhancements exceeding 200 %. Moreover, cluster arrangements of VACHTs benefit from increased power output due to enhanced blockage effects, with co-rotating VACHTs exhibiting higher gains (13.65 %) compared to counter-rotating VACHTs (9.84 %). The lengths of both VACHTs’ wake are similar to H-type turbines, indicating that the compactness of the VACHT cluster remains unaffected.

A hybrid genetic-fuzzy ant colony optimization algorithm for automatic K-means clustering in urban global positioning system

This paper introduces an innovative automatic K-means clustering algorithm, namely HGA-FACO, which seamlessly integrates the noise algorithm, Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Adaptive Fuzzy System (AFS). The rationale behind the HGA-FACO algorithm is to mitigate the shortcomings of traditional K-means, particularly the reliance on pre-determined cluster centers and the need for specifying the number of clusters in advance. By optimizing the search strategy, HGA-FACO efficiently circumvents local optima and effectively explores the global optimal solution, resulting in more accurate and stable clustering outcomes. To validate the superiority of the HGA-FACO over conventional K-Means Clustering (KMeans) and other intelligent clustering approaches such as ACO-KMeans, GA-KMeans (GAK), particle swarm optimization KMeans (PSOK), and ACO-GAK, we conducted comprehensive experiments on taxi Global Positioning System (GPS) datasets sourced from four distinct cities. Employing rigorous evaluation metrics including Silhouette Coefficient (SC), Partition Coefficient (PBM), Davies-Bouldin Index (DBI), and Sum of Squared Errors (SSE), the experimental results convincingly demonstrate that the HGA-FACO significantly outperforms its counterparts across all metrics, highlighting its exceptional performance in clustering effectiveness and compactness. While the HGA-FACO faces challenges related to computational complexity and the necessity for initial parameter tuning, its performance limitations on small-sized or unevenly distributed datasets are acknowledged. Nevertheless, the algorithm's advancements in the field of clustering algorithms are undeniable and hold immense potential for practical applications, notably in city hotspot identification.

Employing Leaf Removal and Spraying Gibberellic Acid (GA3) to Reduce Cluster Compactness Coefficient and Enhance Chemical and Physical Properties of Thompson seedless H4 Grapevines.

Employing Leaf Removal and Spraying Gibberellic Acid (GA3) to Reduce Cluster Compactness Coefficient and Enhance Chemical and Physical Properties of Thompson seedless H4 Grapevines.

Joint time-vertex linear canonical transform

The emergence of graph signal processing (GSP) has spurred a deep interest in signals that naturally reside on irregularly structured data kernels, such as those found in social, transportation, and sensor networks. Recently, concepts and applications related to time-varying graph signal analysis have matured, linking temporal signal processing techniques with innovative tools in GSP. In this paper, similar to the extension of the graph fractional Fourier transform to the graph linear canonical transform, we define the joint time-vertex linear canonical transform (JLCT) and its properties. This transformation extends the joint time-vertex Fourier transform (JFT) and fractional Fourier transform (JFRFT), broadening the Fourier analysis in both time and vertex domains into the domain of the linear canonical transform (LCT). This offers an enhanced set of the LCT analysis tools for joint time-vertex processing. Applications of the JLCT in dynamic mesh denoising, clustering, and energy compactness demonstrate that JLCT can enhance regression and learning tasks, and can refine and improve the performance of the JFT and the JFRFT.

Capacitated Clustering via Majorization-Minimization and Collaborative Neurodynamic Optimization.

This paper addresses capacitated clustering based on majorization-minimization and collaborative neurodynamic optimization (CNO). Capacitated clustering is formulated as a combinatorial optimization problem. Its objective function consists of fractional terms with intra-cluster similarities in their numerators and cluster cardinalities in their denominators as normalized cluster compactness measures. To obviate the difficulty in optimizing the objective function with factional terms, the combinatorial optimization problem is reformulated as an iteratively reweighted quadratic unconstrained binary optimization problem with a surrogate function and a penalty function in a majorization-minimization framework. A clustering algorithm is developed based on CNO for solving the reformulated problem. It employs multiple Boltzmann machines operating concurrently for local searches and a particle swarm optimization rule for repositioning neuronal states upon their local convergence. Experimental results on ten benchmark datasets are elaborated to demonstrate the superior clustering performance of the proposed approaches against seven baseline algorithms in terms of 21 internal cluster validity criteria.

Study on cluster centers optimization of max-min distance k-means clustering algorithm based on inter-cluster separation measure

Clustering is an essential unsupervised technique when category information is not available. Although K-means and Max-min distance K-means clustering algorithms are widely used, they have some disadvantages such as dependence on the initial centers, sensitivity to outliers caused by using only distance as the clustering criterion. To overcome the problems, this paper proposes SMM-K-means algorithm which overcomes the dependence on the initial cluster centers and the initial number of clusters and the sensitivity to the outliers. First, the initial value K of the optimal cluster number is determined by the elbow method, and K-means is used for initial clustering. A new inter-cluster separation measure is then constructed based on the idea of q-nearest neighbors, which is constructed by comprehensive considering the separation between clusters and the distribution compactness of clusters themselves. Finally, the two sample points with highest degree of separation are brought into Max-min distance K-means algorithm as new initial centers for clustering. The definite determining method of cluster centers eliminates the complicated iterative calculation, and the construction of inter-cluster separation measure overcomes the sensitivity of clustering results to noise points and isolated points, and has good applicability and generalization. In addition, this algorithm is not limited by the shape and size of the clusters and has better flexibility. The experimental results show that the SMM-K-means algorithm has higher CH values, resulting in a better clustering effect and stability.

Robust and compact maximum margin clustering for high-dimensional data

In the field of machine learning, clustering has become an increasingly popular research topic due to its critical importance. Many clustering algorithms have been proposed utilizing a variety of approaches. This study focuses on clustering of high-dimensional data using the maximum margin clustering approach. In this paper, two methods are introduced: The first method employs the classical maximum margin clustering approach, which separates data into two clusters with the greatest margin between them. The second method takes cluster compactness into account and searches for two parallel hyperplanes that best fit to the cluster samples while also being as far apart from each other as possible. Additionally, robust variants of these clustering methods are introduced to handle outliers and noise within the data samples. The stochastic gradient algorithm is used to solve the resulting optimization problems, enabling all proposed clustering methods to scale well with large-scale data. Experimental results demonstrate that the proposed methods are more effective than existing maximum margin clustering methods, particularly in high-dimensional clustering problems, highlighting the efficacy of the proposed methods.

ECG-free cine MRI with data-driven clustering of cardiac motion for quantification of ventricular function.

Despite the widespread use of cine MRI for evaluation of cardiac function, existing real-time methods do not easily enable quantification of ventricular function. Moreover, segmented cine MRI assumes periodicity of cardiac motion. We aim to develop a self-gated, cine MRI acquisition scheme with data-driven cluster-based binning of cardiac motion. A Cartesian golden-step balanced steady-state free precession sequence with sorted k-space ordering was designed. Image data were acquired with breath-holding. Principal component analysis and k-means clustering were used for binning of cardiac phases. Cluster compactness in the time dimension was assessed using temporal variability, and dispersion in the spatial dimension was assessed using the Caliński-Harabasz index. The proposed and the reference electrocardiogram (ECG)-gated cine methods were compared using a four-point image quality score, SNR and CNR values, and Bland-Altman analyses of ventricular function. A total of 10 subjects with sinus rhythm and 8 subjects with arrhythmias underwent cardiac MRI at 3.0 T. The temporal variability was 45.6 ms (cluster) versus 24.6 ms (ECG-based) (p< 0.001), and the Caliński-Harabasz index was 59.1 ± 9.1 (cluster) versus 22.0 ± 7.1 (ECG based) (p< 0.001). In subjects with sinus rhythm, 100% of the end-systolic and end-diastolic images from both the cluster and reference approach received the highest image quality score of 4. Relative to the reference cine images, the cluster-based multiphase (cine) image quality consistently received a one-point lower score (p< 0.05), whereas the SNR and CNR values were not significantly different (p= 0.20). In cases with arrhythmias, 97.9% of the end-systolic and end-diastolic images from the cluster approach received an image quality score of 3 or more. The mean bias values for biventricular ejection fraction and volumes derived from the cluster approach versus reference cine were negligible. ECG-free cine cardiac MRI with data-driven clustering for binning of cardiac motion is feasible and enables quantification of cardiac function.

Influence of Grapevine Cultivar on Population Levels of Lobesia botrana (Lepidoptera: Tortricidae) and Effectiveness of Insecticides in Controlling This Pest

The European grapevine moth, Lobesia botrana (Denis and Schiffermüller) (Lepidoptera: Tortricidae), is the most critical pest of vineyards. In the present study, pheromone-baited traps were applied in 2021 and 2022 to monitor the moth population dynamics and to determine the number of L. botrana generations. The number of eggs and larvae was also counted in four vineyards with Askari, Yaghooti, Keshmeshi, and Fakhri cultivars. Moreover, the morphological properties of clusters were evaluated in different grape cultivars to find out the susceptible cultivar to L. botrana. In 2022, different insecticides were used in the Askari cultivar vineyard, and larval damage level was assessed. Three generations were recorded in all vineyards each year. The population of males was not affected by the cultivar. In contrast, the population density of eggs and larvae was significantly higher in Yaghooti than in other tested cultivars. It could be attributed to the cluster compactness and thin skin of berries in Yaghooti, which makes it more susceptible to L. botrana infestations. In contrast, the lowest eggs and larval population density was reported in the Fakhri cultivar indicating the tolerance of this cultivar compared to the other tested cultivars. The field trial showed that the application of insecticides in the second and third generations reduced the damage level of L. botrana. The rotation of insecticides with different modes of action in consecutive generations of L. botrana can be used to reduce damage levels.

The Impact of Bud Load on Berry Quality, Yield, and Cluster Compactness in H4 Strain Grapevines

The Thompson Seedless (Sultanina) seedless variety of grapes is an important crop in Egypt, both for local consumption and export. In recent years, the H4 strain of this grape variety has gained popularity due to its high productivity. However, a drawback of this strain is that the grape clusters become densely packed, resulting in small berries and reduced overall quality. A study was conducted to investigate the impact of pruning severity and bud load on the growth, yield, and quality of H4 grapes. The study included several different treatments, namely T1 Control: (10 canes × 12 buds), T2: (8 canes × 6 buds), T3: (8 canes × 8 buds), T4: (8 canes × 10 buds), T5: (8 canes × 12 buds), T6: (6 canes × 13 buds), T7: (8 canes × 13 buds), T8: (10 canes × 13 buds), and T9: (12 canes × 13 buds). The behavior of the buds, vegetative growth, yield, and quality of the grapes were analyzed. The results showed that pruning of the H4 strain with either 8 canes and 10–12 buds per cane, or 6 canes with 13 buds, produced the best results in terms of managing excessive fruit production, achieving a balance between vegetative growth and yield, and improving the physical and chemical characteristics of the grape clusters and berries. Overall, maintaining 8 canes with 10–12 buds per cane or 6 canes with 13 buds is recommended for obtaining maximum crop yield and quality.

Image segmentation by selecting eigenvectors based on extended information entropy

AbstractFor spectral clustering algorithm, the quality of eigenvectors of graph affinity matrix is very important for the clustering result. So, how to obtain high‐quality eigenvectors is crucial. In this paper, the authors aim to propose some new measurement methods to evaluate each eigenvector of affinity matrix for spectral selection. Based on extended information entropy, three criteria, i.e. Spectral Distinguishability (SD), Spectral Distinguishability Validity (SDV) and pectral Distinguishability ‐Degree (SDD), are defined respectively. The compactness of clusters for each eigenvector is measured by SD; SDV is used to remove the inefficient eigenvectors for clustering; SDD is used to evaluate the contribution of eigenvectors to clustering and is exploited to build a selective spectral ensemble scheme. To indicate the merits of the authors’ algorithm, the authors consider varied artificial data and natural images, including Berkeley image segmentation data set as benchmark data set. The authors’ simulation results confirm the superior performance of the proposed method in developing spectral clustering compared to conventional clustering methods and recent eigenvectors‐selection‐based algorithms.

Ultrasonic sensor-based automatic control volume sprayer for pesticides and growth regulators application in vineyards

A study was conducted to design a sensor-based control volume sprayer for vineyard cultivation. The study aimed to develop cluster-specific pesticide and growth regulator application systems with improved penetration and minimum off-target losses, thus saving chemicals and overall cost. For the determination of design values of the control volume unit, physical properties (cluster length, width, and level of compactness) of grape clusters from seventeen major genotypes selected from vineyard research farm ICAR-IARI New Delhi were studied. For the design of the grape cluster detection unit, two different types of sensors, i.e., ultrasonic sensor (HC SR-04) and infrared proximity sensor (B0115NCT4U), were evaluated for their ability to detect the target (grape cluster) and the response time. The design of the spraying unit for the control volume sprayer also involved selecting suitable nozzles and operational parameters. The developed sprayer was evaluated for application of plant growth hormone (Gibberellic acid), and the performance was assessed in terms of droplet size, chemical application per cluster, cluster and berry growth characteristics, sprayer application rate, uniformity coefficient, and application time. The data on the physical properties of grape clusters revealed a maximum cluster length and width of 24.00 ± 0.91 cm and 18.80 ± 0.15 cm, respectively. The ultrasonic sensor had a comparatively higher sensing range, beam angle, and lower price than the infrared sensor. The spray uniformity for the hollow cone and flat fan nozzle varied significantly for the operating pressure range of 2–4.5 kg.cm−2. The developed sprayer consisted of a 3D printed control volume unit, ultrasonic sensor, two flat fan nozzles, microcontroller (Atmega 328P), relay switch, voltage converter, spray tank of 16-liter capacity, and 12 VDC battery. In case of developed sensor-based control volume sprayer, 30% and 35.48% saving in chemical use and application time, respectively, were observed compared to the conventional dipping method. The PGR application method has significant interaction with cluster length and berry growth as the main effect; however, failed to show a significant interaction with cluster width. Maximum cluster growth (16.89 ± 1.72 mm) was observed in PGR application with a developed sprayer.

Prediction of the optical properties of TiO2 suspension by coupling the classical particle agglomeration model with the discrete dipole approximation algorithm and experimental verification

Nanofluids are widely used in the field of energy and environment. However, the optical properties of such multiphase suspension are quite complex. Even for the well-studied TiO2 suspension, theoretical investigation of its optical parameters is quite rare, which remarkably restricts the extension of their application especially in the field of solar energy and optofluidics. Therefore, in this paper, the optical properties of TiO2 suspension were studied by coupling the classical particle agglomeration model and the discrete dipole approximation algorithm. The absorption and scattering coefficients of TiO2 aqueous suspension have been calculated and verified by experimental results. The effects of cluster compactness, size and other factors on the phase function were analyzed. Our work can also provide a reference for the study of optical properties of other nanofluids.

Fabrication-Aware Joint Clustering in Freeform Space-Frames

The geometrical variability in the joints of large-scale, doubly-curved space-frame structures can have a substantial impact on the time and cost of their construction. This paper proposes a novel framework to assess the construction complexity of space-frame structures as a factor of the geometrical variability and fabrication of their joints, to promote the informed design of the fabrication process. The k-means algorithm was used to cluster space-frame joints into fabrication batches, providing an overview of the variability distribution. A novel initialisation method was developed that allows the algorithm to adapt to project-specific inputs, substantially improving cluster compactness. Overlaying the clustering results with the properties of different fabrication processes provides an accurate estimation of the construction complexity of alternative fabrication options. The method was applied to a large-scale case study to demonstrate the benefits in practice. Alternative fabrication scenarios were assessed in the early stages of the design development, leading to the informed design of the fabrication process and hence to the efficient construction of large-scale, complex structures.

Deep fiber clustering: Anatomically informed fiber clustering with self-supervised deep learning for fast and effective tractography parcellation

White matter fiber clustering is an important strategy for white matter parcellation, which enables quantitative analysis of brain connections in health and disease. In combination with expert neuroanatomical labeling, data-driven white matter fiber clustering is a powerful tool for creating atlases that can model white matter anatomy across individuals. While widely used fiber clustering approaches have shown good performance using classical unsupervised machine learning techniques, recent advances in deep learning reveal a promising direction toward fast and effective fiber clustering. In this work, we propose a novel deep learning framework for white matter fiber clustering, Deep Fiber Clustering (DFC), which solves the unsupervised clustering problem as a self-supervised learning task with a domain-specific pretext task to predict pairwise fiber distances. This process learns a high-dimensional embedding feature representation for each fiber, regardless of the order of fiber points reconstructed during tractography. We design a novel network architecture that represents input fibers as point clouds and allows the incorporation of additional sources of input information from gray matter parcellation. Thus, DFC makes use of combined information about white matter fiber geometry and gray matter anatomy to improve the anatomical coherence of fiber clusters. In addition, DFC conducts outlier removal naturally by rejecting fibers with low cluster assignment probability. We evaluate DFC on three independently acquired cohorts, including data from 220 individuals across genders, ages (young and elderly adults), and different health conditions (healthy control and multiple neuropsychiatric disorders). We compare DFC to several state-of-the-art white matter fiber clustering algorithms. Experimental results demonstrate superior performance of DFC in terms of cluster compactness, generalization ability, anatomical coherence, and computational efficiency.

The Association between Carbon Emission and Urban Spatial Form—A Study of Zhuhai, China

Research on carbon emission is an important basis for solving global climate problems, and it is also one of the ways to quantitatively assess the impact of human activities. Cities are one of the main bearing spaces of human activities, and reasonable urban form is conducive to reducing energy consumption in human activities. This paper takes 50 clusters within Zhuhai, China, as the research object, quantifies the landscape spatial form index and social spatial form index of each cluster and establishes the index set of urban spatial form, so as to analyze the influence of different urban spatial form index on carbon emission. The main conclusions are as follows: (1) From the spatial distribution of each index, the urban cluster size and residential building area of each cluster in Zhuhai are generally large, and the distribution is basically consistent with that of densely populated areas. The urban clusters with high dominance are mainly located in the main urban area of Xiangzhou District, and the urban compactness, dispersion and industrial building area are generally high in the west and low in the east. (2) The size of urban clusters, industrial building area and residential building area have a strong promoting effect on carbon emission, while the compactness, dispersion and dominance of urban clusters have a strong inhibiting effect on carbon emission. (3) Based on the above conclusions, the low-carbon urban spatial form optimization strategy should be proposed from three aspects: urban development boundary control, promoting industrial structure transformation and compact urban development.

GA3 and Hand Thinning Improves Physical, Chemical Characteristics, Yield and Decrease Bunch Compactness of Sultanina Grapevines (Vitis vinifera L.)

Nowadays, the Sultanina grapevines H4 strain has become widely cultivated because of its high productivity. However, this Sultanina variety is suffering from clusters with small berries, in addition highly compacted berries, thus negatively affecting the quality of bunch berries. A field experiment was carried out during the two successive seasons on Five years old Sultanina (H4 strain) grapevines grafted onto freedom rootstock grown in a private orchard located in El-Khatatba region, Minufyia Governorate, with coordinates of 30°21′ N 30°49′ E. The investigation was designed to throw light on the effect of hand thinning at levels 0, 25%, and or 50% of cluster shoulders was removed in addition to spray with 0, 20, 30, and 40 ppm GA3 for berry sizing. In addation, the first application was the hand thinning treatment, which was used on the second and third week of May for the first and the second seasons. The second application was the berry sizing treatment, applied when the berries were at 6–7 mm diameter (on the 3rd and 4th week of May for the first and the second seasons. Results indicated that the total chlorophyll content of leaf decreases when the hand thinning levels or GA3 concentration increases, while hand thinning and GA3 treatments were higher than control only concerning pruning weight. Moreover, the results showed that the first level of hand thinning treatments recorded higher values for the cluster weight than the higher level of hand thinning treatments with all the concentrations of GA3 used for sizing. The highest value was with T2. The overperformance of T2 for the vine yield and the rest of the treatments at the lower level of hand thinning were better than those at the higher level of hand thinning. In conclusion, hand thinning at 50% level and sizing with GA3 at concentration 30 ppm reduced the cluster compactness and improved the berry weight and firmness. Therefore, it could be recommended to get the most suitable yield and quality of Sultanina grapevines.