Nearest-neighbor Method Research Articles

Development and Application of Adaptive Evaluation System for TBM Tunneling Based on Case-Based Reasoning

The excavation adaptability evaluation decision process for the TBM (tunnel boring machine) in long and large tunnels under complex geological conditions is an uncertain and fuzzy problem affected by multiple factors. Aiming to shift the status quo of TBMs tunneling adaptability evaluation and the ineffective use of related accident cases, the TBM tunneling adaptability evaluation method based on case-based reasoning is proposed and researched. The case-based reasoning method can use existing engineering experience and knowledge to answer newly encountered problems, and can accurately and efficiently evaluate the adaptability of TBM tunneling. Based on the nearest neighbor method, this paper establishes the calculation formula of TBM tunneling adaptability similarity. Based on the statistical analysis method, the evaluation indicators that play a relatively important role in the system are selected, and the TBM tunneling adaptability evaluation index system is constructed. The analytic hierarchy process (AHP) is used to determine the weight of the evaluation indicators at each level. According to the characteristics of TBM tunneling adaptability evaluation, the overall design of the case-based reasoning-based TBM tunneling adaptability evaluation decision system CBR-TBMEAEDS (case-based reasoning-TBM excavation adaptive evaluation decision system) is proposed, and the TBM tunneling adaptability evaluation case is expressed The case-based reasoning method and modification method were designed, and CBR-TBMEAEDS was developed. The system can be used to evaluate the adaptability of TBM to the constructed case library, and the evaluation results are consistent with the actual situation.

An efficient indexing technique for billion-scale nearest neighbor search

An efficient indexing technique for billion-scale nearest neighbor search

Privacy-Preserving Public Route Planning Based on Passenger Capacity

Precise route planning needs huge amounts of trajectory data recorded in multimedia devices. The data, including each user’s location privacy, are stored as cipher text. The ability to plan routes on an encrypted trajectory database is an urgent necessity. In this paper, in order to plan a public route while protecting privacy, we design a hybrid encrypted random bloom filter (RBF) tree on encrypted databases, named the encrypted random bloom filter (eRBF) tree, which supports pruning and a secure, fast k nearest neighbor search. Based on the encrypted random bloom filter tree and secure computation of distance, we first propose a reverse k nearest neighbor trajectory search on encrypted databases (RkNNToE). It returns all transitions, in which each takes the query trajectory as one of its k nearest neighbor trajectories on the encrypted database. The results can be the indicator of a new route’s capacity in route planning. The security of the trajectory and query is proven via the simulation proof technique. When the number of points in the trajectory database and transition database are 1174 and 18,670, respectively, the time cost of an R2NNToE is about 1200 s.

Pulmonary Vein Isolation With and Without Posterior Wall Isolation in Paroxysmal Atrial Fibrillation: IMPPROVE-PAF Trial

BackgroundPrior studies have demonstrated clinical benefits associated with cryoballoon pulmonary vein isolation (PVI) and concomitant posterior wall isolation (PWI) in patients with persistent atrial fibrillation (AF). However, the role for this approach in patients with paroxysmal atrial fibrillation (PAF) remains unclear. ObjectivesThis study investigated the acute and long-term outcomes of PVI vs PVI+PWI using cryoballoon in patients with symptomatic PAF. MethodsThis retrospective study (NCT05296824) examined the outcomes of cryoballoon PVI (n = 1,342) vs cryoballoon PVI+PWI (n = 442) in patients with symptomatic PAF during long-term follow-up. Using the nearest-neighbor method, a 1:1 matched sample of patients receiving PVI alone and PVI+PWI was created. ResultsThe matched cohort consisted of 320 patients (PVI: n = 160; PVI+PWI: n = 160). PVI+PWI was associated with longer cryoablation (23 ± 10 minutes vs 42 ± 11 minutes; P < 0.001) and procedure times (103 ± 24 minutes vs 127 ± 14 minutes; P < 0.001). In 39 (24.4%) of 160 patients, adjunct radiofrequency ablation was required for PVI+PWI. Adverse event rates were similar (PVI 3.8% vs PVI+PWI 1.9%; P = 0.31). Though there were no differences at 12 months, freedom from all atrial arrhythmias (67.5% vs 45.0%; P < 0.001) and AF (75.6% vs 55.0%; P < 0.001) were significantly greater with PVI+PWI vs PVI alone at 39 ± 9 months of follow-up. PVI+PWI was also associated with reduced long-term need for cardioversion (16.9% vs 27.5%; P = 0.02) and repeat catheter ablation (11.9% vs 26.3%; P = 0.001), and emerged as the only significant predictor of freedom from recurrent AF (HR: 2.79; 95% CI: 1.64-4.74; P < 0.001). ConclusionsCompared with cryoballoon PVI, cryoballoon PVI+PWI appears to be associated with greater freedom from recurrent atrial arrhythmias and AF in patients with PAF during long-term follow-up >3 years.

Robust multi-view non-negative matrix factorization with adaptive graph and diversity constraints

Multi-view clustering (MVC) has received extensive attention due to its efficient processing of high-dimensional data. Most of the existing multi-view clustering methods are based on non-negative matrix factorization (NMF), which can achieve dimensionality reduction and interpretable representation. However, there are following issues in the existing researches: (1) The existing methods based on NMF using Frobenius norm are sensitive to noises and outliers. (2) Many methods only use the information shared by multi-view data, while ignoring the diverse information between views. (3) The data graph constructed by the conventional K Nearest Neighbors (KNN) method may misclassify neighbors and degrade the clustering performance. To address the above problems, we propose a novel robust multi-view clustering method. Specifically, l2,1-norm is introduced to measure the factorization error to improve the robustness of NMF. Additionally, a diversity constraint is utilized to learn the diverse relationship of multi-view data, and an adaptive graph method via information entropy is designed to overcome the shortcomings of misclassifying neighbors. Finally, an iterative updating algorithm is developed to solve the optimization model, which can make the objective function monotonically non-increasing. The effectiveness of the proposed method is substantiated by comparing with eleven state-of-the-art methods on five real-world and four synthetic multi-view datasets for clustering tasks.

A link prediction method based on topological nearest-neighbors similarity in directed networks

Link prediction is a fundamental and key field in complex network research, and some scholars have conducted various studies in this field. However, most of the existing link prediction methods neither consider the direction of the network edge nor make full use of the information of the network node. This paper proposes a topological nearest-neighbors similarity method in a directed network to solve this problem. Firstly, this study improved the Sorensen index in directed networks, and its variants also are proposed. Secondly, the matrix form of each basic index is expressed using matrix algebra. Then, based on the idea of GLHN(Global Leicht Holme Newman) similarity index, the nearest-neighbors topology of each basic index is derived to obtain the topological nearest-neighbors similarity index. Finally, the proposed method is validated empirically using multiple real directed network datasets. Experiments verify the superiority of the proposed method by comprehensively using three evaluation metrics compared with the benchmark indices, including lower error, higher accuracy and stronger robustness.

A Machine Learning Approach for Transmitted Power Estimation in Power Beaming Systems

Currently, predictive machine learning methods are used in many areas of life — from traffic predictions to medical diagnosis. Recently, these approaches also appeared in atmoposheric studies, first of all, for estimation of turbulence parameters, these tasks are necessary for a qualitative solution of atmospheric optical communication issues. The purpose of this work is to show the possibility and prospects of using machine learning algorithms for estimation transmitted power in power beaming systems in real time under changing atmospheric conditions. Experimental data were collected over several months on long atmospheric experimental setup, among gathered data there are such meteorological parameters as pressure, temperatures, wind speed, humidity, dew point, wind direction, solar flux. The data was collected for several locations. The power of the incident radiation was estimated from the voltage on the photovoltaic receiver. The nearest neighbors method, gradient boosting and neural networks were used as machine learning algorithms, the algorithms were compared with each other in terms of the average absolute error (MAPE) and the coefficient of determination (R2). The analysis of the results showed a good predictive ability of all models and potential of using even on the basis of simple meteorological measurements. The use of traditional methods requires much more complex measurements, such as scintillometry, or empirical approximations are used. Machine learning makes it possible to get results with only integral meteorological parameters and shows good accuracy for arbitrary conditions. Gradient boosting with R2 0.951 and MAPE 0.020 on all data was chosen as the best model. The results of this model was interpreted using the SHAP method, the dependence of the result on the input data is consistent with expectations.

Weather Parameter Analysis Using Interpolation Methods

One of the key components in minimizing losses has been using rainfall forecasts at various temporal and spatial scales by businesses, governments, and farmers. Forecasts for other atmospheric characteristics are also important for agriculture. To maintain accurate weather forecasts, atmospheric procedures must be understood, measurements must be improved, and research and development must be continued. India’s Ministry of Earth Science has many programs and agencies, such as the India Meteorological Department (IMD), that strive to enhance forecasting. The efforts have culminated in fairly reliable predictions of rainfall patterns. This research aims to enhance prediction accuracy in different geographical areas ranging from weather subdivisions to agro-climate areas. We have used data that were obtained from the IMD, Pune. In the preprocessing technique session, we transformed data from the txt format to the csv format. We have used Matlab software for data analysis. Interpolation techniques have introduced detailed experimental position forecasts for the Marathwada region of Maharashtra, India. The linear interpolation method produces better results than a cubic, nearest neighbor, spline, shape-preserving, and PCHIP interpolation methods. The accuracy of linear interpolation methods is 90.92%. The accuracy of the cubic and PCHIP interpolation techniques is 90.76%, the accuracy of the shape-preserving and spline interpolation methods is 88.93%, and the accuracy of the nearest neighbor is 90.70%. Normal, heavy, and low rainfall are only 30–35%, 20–25%, and 35–40%, respectively. Marathwada region rainfall increased by 40 mm from 2021 to 2020. The annual temperature is 36–37°C in the Marathwada region, due to this Marathwada region falls under drought. So there is a need to change the crop pattern in the Marathwada region like temperature tolerant crops.

Association of late tricuspid regurgitation progression and intracardiac lead types in patients after primary implantation of cardiac electronic devices

Aim. To evaluate the factors and degree of progression of tricuspid regurgitation (TR) depending on the type of implanted electrode in patients after primary implantation of a cardiac implantable electronic device (CIED) in the late postoperative period.Methods. Case histories of 674 patients who underwent primary implantation of an implantable cardioverterdefibrillator (ICD) for the secondary prevention of sudden cardiac death for the period from 2009 to 2019 were analyzed and 75 case histories were selected. Using similar criteria, a comparison group was selected, who were implanted with a permanent pacemaker (PM) for sinus node dysfunction. For maximum comparability of the main and reference groups, pseudo-randomization was used using the nearest neighbor search method; 68 pairs were formed, comparable in terms of the main factors. Multiple logistic regression was used to analyze the predictors of progression of TR in the postoperative period.Results. In the preoperative period in the ICD group, 34 patients (50%) had a mild degree of TR, moderate and severe TR occurred in an equal number of cases 2 patients (2.9%) each. In the PM group, mild TR was detected in 24 cases (35.3%), moderate in 2.9% (n=2) of cases, respectively. In the postoperative period, predominantly mild TR was determined in both groups (in the ICD group in 67.6% (n=46), in the PM group in 48 patients (70.6%)).In the analysis of the general group of patients with implanted CIEDs (n=136), in the late postoperative period, an increase in the volume of both atria and the degree of TR was revealed. Left ventricular ejection fraction and end-diastolic volume didn’t have significant dynamics. All echocardiographic data obtained in the postoperative period were within the normal range.According to the results of multiple logistic regression, a history of coronary heart disease (CHD) has the greatest predictive power (hazard ratio 4.170; 95% confidence interval 1.751-9.933, p=0.001).Conclusion. TR in patients after primary implantation of a CIED in the late postoperative period progresses slightly, regardless of the type of right ventricular electrode. A long history of CHD is associated with the greatest risk of progression to tricuspid valve insufficiency.

Spatial scale applicability of 4-Scale geometrical optics model canopy reflectance simulation.

Accurately clarifying the applicable spatial scale of 4-Scale model is conducive to improving the accuracy of its application in canopy reflectance simulation of different vegetation types, and to further improving the inversion accuracy of leaf area index, canopy density, and other parameters. Two forest plots (one for broad-leaved forest and one for mixed forest) with each area of 100 m×100 m in Maoershan Experimental Forest Farm, Shangzhi, Heilongjiang, were divided into the spatial scales of 10, 20, 30, 40 and 50 m, respectively. The 4-Scale model was used to simulate forest canopy reflectance. Local mean method, the nearest neighbor method, bilinear interpolation method, and cubic convolution method were used to convert Sentinel-2 images with spatial resolution of 10 m to other scales, with the results being evaluated. The simulated canopy reflectance and remote sensing pixel reflectance were compared and analyzed. The spatial scale of mixed forest and broad-leaved forest suitable for high-precision inversion parameters of 4-Scale model was determined. The results showed that the 4-Scale model underestimated the pixel forest canopy reflectance as a whole. The canopy reflectance of mixed forest and broad-leaved forest had the worst simulation effect at the 20 m scale. Both the root mean square error (RMSE) and the mean absolute error from (MAE) of red and near-infrared band were large. When the scale was >20 m, the simulation effect became better. The applicability of the model was the best when the mixed forest was 40 m and the broad-leaved forest was 30 m. The mean and standard deviation of the reflectance difference between the simulated value and the remote sensing pixel were the minimum in the red and near near-infrared bands, with the minimum RMSE and MAE. The simulation results of mixed forest and broad-leaved forest at 10 m scale were not stable, the rule of mean and standard deviation was inconsistent, and the difference between RMSE and MAE was large under the same band.

Analysis of braid manifolds using uniform manifold approximation and projection

Autonomous sonar target recognition has been an ongoing challenge in the underwater signal processing community for many decades. Aspect-dependent target scattering and waveguide propagation causes target features to overlap, which increases the difficulty of constructing feature dictionaries. Current machine learning algorithms can suffer from an inability to track morphing target-specific features. Other ”black-box” machine learning algorithms produce results that are not explainable. We seek to extend previous work on creating feature representations using braid manifolds. Specifically, we use Uniform Manifold Approximation and Projection (UMAP) as a dimensionality reduction technique to search for underlying features that lay on a manifold. UMAP is an unsupervised learning algorithm that is built upon Riemannian geometry and fuzzy simplicial sets. The low dimensional representation embedding that UMAP produces is computed using both stochastic approximate nearest neighbor search and stochastic gradient descent with negative sampling. The performance of UMAP embeddings will be compared against other common dimension reduction algorithms and evaluated using various classifiers. [The authors would like to thank the Office of Naval Research for funding under Grant No. N00014-21-1-2420 and the University of Iowa.]

A Novel Fault Detection Model Based on Vector Quantization Sparse Autoencoder for Nonlinear Complex Systems

To solve the problem of nonlinear factors in the fault detection process of complex systems, this article proposes a fault detection model based on vector quantization sparse autoencoder. First, a feature extraction model, which consists of a self-normalizing convolutional autoencoder module, a vector quantization module, a gradient module, and a loss module, is developed. The first module employs self-normalizing convolutional layers with good stability and generalization ability to extract the nonlinear structural features of complex systems. A nearest neighbor search strategy is implemented in the vector quantization module to further mine the nonlinear information. The gradient module adopts a straight-through estimation technique to improve the training efficiency. Sparse constraints are introduced into the loss module to obtain the essential features and enhance interpretability. Thereafter, a construction rule based on local Mahalanobis distance and K nearest neighbors is designed to calculate K Mahalanobis neighbor metrics that depend on the sparse features obtained by the feature extraction model. A comprehensive statistic for fault detection is constructed to accurately track the operating status of complex systems by combining the loss metric and the K Mahalanobis neighbor metric. Finally, the threshold of the fault detection statistics is determined by modeling the generalized extreme value distribution. Three case studies, a numerical simulation, the Tennessee Eastman benchmark process, and a typical circuit system, are adopted to demonstrate the effectiveness and merits of the proposed fault detection model.

Comparison of Machine Learning Algorithms for Natural Gas Identification with Mixed Potential Electrochemical Sensor Arrays

Mixed-potential electrochemical sensor arrays consisting of indium tin oxide (ITO), La0.87Sr0.13CrO3, Au, and Pt electrodes can detect the leaks from natural gas infrastructure. Algorithms are needed to correctly identify natural gas sources from background natural and anthropogenic sources such as wetlands or agriculture. We report for the first time a comparison of several machine learning methods for mixture identification in the context of natural gas emissions monitoring by mixed potential sensor arrays. Random Forest, Artificial Neural Network, and Nearest Neighbor methods successfully classified air mixtures containing only CH4, two types of natural gas simulants, and CH4+NH3 with >98% identification accuracy. The model complexity of these methods were optimized and the degree of robustness against overfitting was determined. Finally, these methods are benchmarked on both desktop PC and single-board computer hardware to simulate their application in a portable internet-of-things sensor package. The combined results show that the random forest method is the preferred method for mixture identification with its high accuracy (>98%), robustness against overfitting with increasing model complexity, and had less than 10 ms training time and less than 0.1 ms inference time on single-board computer hardware.

Spatiotemporal Analysis of the Background Seismicity Identified by Different Declustering Methods in Northern Algeria and Its Vicinity

The main purpose of this paper was to, for the first time, analyse the spatiotemporal features of the background seismicity of Northern Algeria and its vicinity, as identified by different declustering methods (specifically: the Gardner and Knopoff, Gruenthal, Uhrhammer, Reasenberg, Nearest Neighbour, and Stochastic Declustering methods). Each declustering method identifies a different declustered catalogue, namely a different subset of the earthquake catalogue that represents the background seismicity, which is usually expected to be a realisation of a homogeneous Poisson process over time, though not necessarily in space. In this study, a statistical analysis was performed to assess whether the background seismicity identified by each declustering method has the spatiotemporal properties typical of such a Poisson process. The main statistical tools of the analysis were the coefficient of variation, the Allan factor, the Markov-modulated Poisson process (also named switched Poisson process with multiple states), the Morisita index, and the L–function. The results obtained for Northern Algeria showed that, in all cases, temporal correlation and spatial clustering were reduced, but not totally eliminated in the declustered catalogues, especially at long time scales. We found that the Stochastic Declustering and Gruenthal methods were the most successful methods in reducing time correlation. For each declustered catalogue, the switched Poisson process with multiple states outperformed the uniform Poisson model, and it was selected as the best model to describe the background seismicity in time. Moreover, for all declustered catalogues, the spatially inhomogeneous Poisson process did not fit properly the spatial distribution of earthquake epicentres. Hence, the assumption of stationary and homogeneous Poisson process, widely used in seismic hazard assessment, was not met by the investigated catalogue, independently from the adopted declustering method. Accounting for the spatiotemporal features of the background seismicity identified in this study is, therefore, a key element towards effective seismic hazard assessment and earthquake forecasting in Algeria and the surrounding area.

Fuzzy Rough Nearest Neighbour Methods for Aspect-Based Sentiment Analysis

Fine-grained sentiment analysis, known as Aspect-Based Sentiment Analysis (ABSA), establishes the polarity of a section of text concerning a particular aspect. Aspect, sentiment, and emotion categorisation are the three steps that make up the configuration of ABSA, which we looked into for the dataset of English reviews. In this work, due to the fuzzy nature of textual data, we investigated machine learning methods based on fuzzy rough sets, which we believe are more interpretable than complex state-of-the-art models. The novelty of this paper is the use of a pipeline that incorporates all three mentioned steps and applies Fuzzy-Rough Nearest Neighbour classification techniques with their extension based on ordered weighted average operators (FRNN-OWA), combined with text embeddings based on transformers. After some improvements in the pipeline’s stages, such as using two separate models for emotion detection, we obtain the correct results for the majority of test instances (up to 81.4%) for all three classification tasks. We consider three different options for the pipeline. In two of them, all three classification tasks are performed consecutively, reducing data at each step to retain only correct predictions, while the third option performs each step independently. This solution allows us to examine the prediction results after each step and spot certain patterns. We used it for an error analysis that enables us, for each test instance, to identify the neighbouring training samples and demonstrate that our methods can extract useful patterns from the data. Finally, we compare our results with another paper that performed the same ABSA classification for the Dutch version of the dataset and conclude that our results are in line with theirs or even slightly better.

A Survey on Deep Hashing Methods

Nearest neighbor search aims at obtaining the samples in the database with the smallest distances from them to the queries, which is a basic task in a range of fields, including computer vision and data mining. Hashing is one of the most widely used methods for its computational and storage efficiency. With the development of deep learning, deep hashing methods show more advantages than traditional methods. In this survey, we detailedly investigate current deep hashing algorithms including deep supervised hashing and deep unsupervised hashing. Specifically, we categorize deep supervised hashing methods into pairwise methods, ranking-based methods, pointwise methods as well as quantization according to how measuring the similarities of the learned hash codes. Moreover, deep unsupervised hashing is categorized into similarity reconstruction-based methods, pseudo-label-based methods, and prediction-free self-supervised learning-based methods based on their semantic learning manners. We also introduce three related important topics including semi-supervised deep hashing, domain adaption deep hashing, and multi-modal deep hashing. Meanwhile, we present some commonly used public datasets and the scheme to measure the performance of deep hashing algorithms. Finally, we discuss some potential research directions in conclusion.

Dimensionality reduction method based on multiple feature-space collaborative discriminative projection for rotor fault diagnosis

At present, the trend for complex and intelligent rotating machinery and equipment, which generate a large amount of high-dimensional and nonlinear fault monitoring data that is difficult to handle, is becoming more and more apparent. This makes the traditional dimensionality reduction algorithms based on point-to-point metrics or a small number of graph-embedding structures lose their utility. To solve this problem, a multiple feature-space collaborative discriminative projection (MFSCDP) algorithm for rotor fault data set dimensionality reduction is proposed. The algorithm first improves the projection metric from sample point to feature space into the median metric in order to achieve the effect of weakening the extrapolation error of the algorithm, and based on this, we propose a sample point-to-point guided nearest-neighbor feature space selection method to improve the construction efficiency of the feature space embedding graph. Then, we use Relief F to indirectly construct the reduced dimensional projection matrix with multiple feature space of collaboration. Finally, the proposed MFSCDP algorithm is used for the dimensionality reduction process of a rotor fault data set. The algorithm’s performance was verified using experimental information from rotor failure simulations of two different structural types. The results show that the algorithm can reduce the difficulty of fault classification and improve the accuracy of identification.

GGNN: Graph-Based GPU Nearest Neighbor Search

Approximate nearest neighbor (ANN) search in high dimensions is an integral part of several computer vision systems and gains importance in deep learning with explicit memory representations. Since PQT, FAISS, and SONG started to leverage the massive parallelism offered by GPUs, GPU-based implementations are a crucial resource for today's state-of-the-art ANN methods. While most of these methods allow for faster queries, less emphasis is devoted to accelerating the construction of the underlying index structures. In this paper, we propose a novel GPU-friendly search structure based on nearest neighbor graphs and information propagation on graphs. Our method is designed to take advantage of GPU architectures to accelerate the hierarchical construction of the index structure and for performing the query. Empirical evaluation shows that GGNN significantly surpasses the state-of-the-art CPU- and GPU-based systems in terms of build-time, accuracy and search speed.

SemanticLoop: Loop Closure With 3D Semantic Graph Matching

Loop closure can effectively correct the accumulated error in robot localization, which plays a critical role in the long-term navigation of the robot. Traditional appearance-based methods rely on local features and are prone to failure in ambiguous environments. On the other hand, object recognition can infer objects' category, pose, and extent. These objects can serve as stable semantic landmarks for viewpoint-independent and non-ambiguous loop closure. However, there is a critical object-level data association problem due to the lack of efficient and robust algorithms. We introduce a novel object-level data association algorithm, which incorporates IoU, instance-level embedding, and detection uncertainty, formulated as a linear assignment problem. Then, we model the objects as TSDF volumes and represent the environment as a 3D graph with semantics and topology. Next, we propose a graph matching-based loop detection based on the reconstructed 3D semantic graphs and correct the accumulated error by aligning the matched objects. Finally, we refine the object poses and camera trajectory in an object-level pose graph optimization. Experimental results show that the proposed object-level data association method significantly outperforms the commonly used nearest neighbor method in accuracy. Our graph matching-based loop closure is more robust to environmental appearance changes than existing appearance-based methods.

QSAR Analysis, Molecular Docking and ADME Studies of Thiobarbituric Acid Derivatives as Thymidine Phosphorylase Inhibitors: A Rational Approach to Anticancer Drug Design by in silico Modelling

Background: Thymidine Phosphorylase (TP) is an imperative target for cancer researchers. In the current research, quantitative structure-activity relationship (QSAR) models were demonstrated to identify new TP inhibitors. Objective: The main objective is to perform a QSAR study on a series of 19 derivatives of thiobarbituric acid and new molecules designed and dock to check potency and efficacy for anticancer activity. Methods: Multiple linear regression analysis (MLR) was used to establish a two-dimensional quantitative structure-activity relationship (2D-QSAR) with regression coefficient values of 0.9781, 0.9513, and 0.9819 for the training set (r2), leave-one-out (LOO) dependent internal regression (q2), and external test set regression (r2 _pred), respectively. Three-dimensional quantitative structure-activity relationship (3DQSAR) model, obtained by using the simulated annealing k nearest neighbour (SA-KNN) method (q2 = 0.7880). Newly designed molecules were subjected to docking studies with 7-deazaxanthine taken as standard. Results: Molecular modelling, structure-based drug design and docking study analysis were performed. The new chemical entities (NCE’s) designed, docked towards targeted receptor and show good results as compared to the standard 7-deazaxanthine. It was found that these molecules bind similar amino acid pocket regions as that of standard. Molecules bind at the active site of TP enzyme involving H bond interactions with shorter distances showed greater affinity. At last, the oral bioavailability and toxic effect were evaluated as absorption, distribution, metabolism, and elimination (ADME) studies by computational means of the Qikprop tool of Schrodinger. Conclusion: One of the most successful and fast-increasing methodologies is molecular modelling. It not only aids in the prediction of specific target compounds but also aids in the cost reduction of valuable substances. QSAR and docking study was performed, and most of the molecules have shown good dock scores. Based on these results, NCE’s for anticancer activity were successfully designed and analysed in this research work which will be helpful for effective drug synthesis with less toxicity in the future. Others: 2D QSAR model was generated by three methods, and the best one was selected for further study. NCEs were planned based on descriptors such as topological, electrostatic, steric, and hydrophobic substitutions around the core.