AU2021103330A4 - Method and apparatus for social distance using clustering - Google Patents

Abstract

METHOD AND APPARATUS FOR SOCIAL DISTANCE USING CLUSTERING Today in modern era safety from COVID-19 is major challenging issue. Social distance one of them making sufficient distance between two people. In this invention we are shows proper social distance between two people by models based Clustering. We are making here cluster for social distance. Clustering is unsupervised learning here machine learning set and will classify the objects into a particular Social distances. They will be represented by cluster, cluster2, cluster3. This invention the efforts of the inventor to ensure the well-being of public health. The proposed framework provides a comprehensive safety from COVID-19 during disastrous circumstances using Clustering. The proposed system ensures that the proper help reaches the people during critical times before it is too late. 1/4 Drawings Cluster 1 0 0 0 Cluster3 Cluster2 000 000t 0 Figure 1:-Social Distance Clusters for People

Description

1/4

Drawings

Cluster 1 0 0

0 Cluster3

Cluster2 000 000t 0

Figure 1:-Social Distance Clusters for People

DESCRIPTION METHOD AND APPARATUS FOR SOCIAL DISTANCE USING CLUSTERING FIELD OF THE INVENTION

[OO]The present invention is about the process of safety with social distance from COVID-19. Making sufficient distance between two people. In this invention we are shows proper social distance between two people by model based Clustering. We making here cluster for social distance. The proposed process provides a comprehensive solution for clusters using accuracy and social distance factors.

BACKGROUND OF THE INVENTION

[002]Refining Initial Points for k-means Clustering Pragmatic ways to deal with grouping utilize an iterative technique (for example K-Means, EM) which joins to one of various nearby minima. It is realized that these iterative strategies are particularly delicate to beginning conditions. We present a strategy for processing a refined beginning condition from a given starting one that depends on an effective procedure for assessing the methods of a circulation. The refined introductory beginning condition permits the iterative calculation to combine to a "superior" nearby least. The technique is relevant to a wide class of grouping calculations for both discrete and nonstop information. We exhibit the utilization of this technique to the famous K-Means clustering calculation and show that refined introductory beginning stages to be sure lead to improved arrangements. Refinement run time is extensively lower than the time needed to group the full information base. The strategy is versatile and can be combined with an adaptable clustering calculation to address the huge scope grouping [1].

[003]Geometric Clustering We proposes an appearance-based picture clustering approach called GGCI (worldwide mathematical grouping for picture). For face pictures taken with fluctuating posture, demeanor, eyes (wearing shades or not) or object pictures under various survey conditions, GGCI utilizes handily estimated nearby metric data to gain proficiency with the basic worldwide math of pictures space, at that point apply the all-inclusive closest neighbor way to deal with bunch pictures. Not quite the same as the standard closest neighbor approach, GGCI considers the thickness around the closest focuses inside groups. Besides, our methodology groups dependent on the geodesic distance measure rather than Euclidean distance measure, which better mirrors the inherent mathematical design of complex implanted in high dimensional picture space. [2].

[004]Clustering and the Continuous k-means Algorithm talked about in the going with article, include datasets so enormous that their immediate control is unreasonable. Some technique for information compression or combination should initially be applied to lessen the size of the dataset without losing the fundamental character of the information. All combination strategies penance some detail; the best techniques are computationally productive and yield results that are, in any event for pragmatic applications, delegate of the first information. Here we present a few broadly utilized calculations that combine information by bunching, or gathering, and afterward present another strategy, the ceaseless k-implies algorithm,* created at the Laboratory explicitly for grouping huge datasets[3].

[005]Hesitant Distance Similarity Measures for Document Clustering This paper presents new approach, Hesitant Distance Similarity Measures for Document Clustering. The proposed Hesitant Distance Similarity Measures approach is based on Fuzzy Hesitant Sets. In this paper we have used fifty Similarity Measures from fl to f50. The steps, Document collection, Text Pre-processing, Feature Selection, Indexing, Clustering Process and Results Analysis are used. Twenty News group data sets are used in the Experiments. The experimental results are evaluated using the Analytical SAS 9.0 Software. The Experimental Results show the proposed approach out performs

[4].

[006]Optimization of geochemical anomaly detection using a novel genetic K-means clustering (GKMC) algorithm Because of the confounded qualities of provincial geochemical information from stream silt because of the intricacy of land highlights, recognition of multi-essential geochemical impressions of mineral stores of interest is a difficult errand. As an approach to address this, a half and half hereditary calculation based method, in particular hereditary K-implies bunching (GKMC) calculation, is proposed here for ideal outline of multi-basic examples (both inconsistency and foundation) in stream residue geochemical information. To do as such, factor investigation and test catchment bowl displaying were combined with GKMC and customary K-implies grouping (TKMC) strategies for distinguishing proof of irregular multi-basic geochemical impressions of stores of porphyry copper and skarn copper in the 1:100,000 scale Varzaghan map sheet, northwest Iran [5].

[007]Cluster: Cluster Analysis Basics and Extensions Genotype x natural collaboration (GxE) can prompt contrasts in execution of genotypes over conditions. GxE examination can be utilized to investigate the steadiness of genotypes and the worth of test areas. We fostered a Rlanguage program (RGxE) that registers univariate security insights, spellbinding measurements, pooled ANOVA, genotype F proportion across area and climate, bunch examination for area, and area connection with normal area execution. Univariate strength measurements determined are relapse slant (bi), deviation from relapse (S2d), Shukla's fluctuation (ai2), S square Wricke's ecovalence (Wi), and Kang's yield soundness (YSi). RGxE is free and planned for use by researchers contemplating execution of polygenic or quantitative characteristics over numerous conditions [6].

[008]The improved performance in terms of efficient clustering the proposed process a algorithm to improve possible social distance measurements based area and/or dimensions is shown in this invention.

OBJECTS OF THE INVENTION

[009]To received set of social distance clusters.

[0010]To have the parameters of social distance clusters 6 feet, less than 6 Feet, More than 6 Feet.

[0011]To use the proposed clustering technique for social distance measurements.

[0012]To show the clusters in different color for social distance this can be used for process of cluster.

[0013]To construct clusters for parameters of social distance with color green, blue and red. Green cluster show good social distance, blue cluster show sufficient social distance and red cluster show less social distance.

SUMMARY OF THE INVENTION

[0014]The present invention is related to process of find out social distance clusters of people. The proposed process has the Clustering is unsupervised learning here machine learning set and will classify the objects into a particular Social distances. They will be represented by cluster, cluster2, cluster3. This invention the efforts of the inventor to ensure the well-being of public health. The proposed framework provides a comprehensive safety from COVID-19 during disastrous circumstances using Clustering. The proposed system ensures that the proper help reaches the people during critical times before it is too late.

BRIEF DESCRIPTION OF DRAWING

[0015]Figure 1is The Model of the showing social distance clusters of people.

[0016]Figure 2 is Social distance for cluster.

[0017]Figure 3 is Social distance for cluster2.

[0018]Figure 4 is Social distance for cluster3.

DETAILED DESCRIPTION OF THE INVENTION

[0019]The proposed design is made showing clusters as Scene of social distance (SOSD) shown in Figure 1. Where social distance between people detection done. Here we are making three clusters cluster (Green), cluster2 (Blue), cluster3 (Red) where people used the mobile phone be smart phone and trace location and calculate social distance between mobile phone.

[0020]Further we calculate the distance and we will call a query to Telecom service provider (TSP), the TSP provided a list of mobile numbers at interval of two -two minutes and with their time, latitude and longitude. The placing position of mobile phones is drawing in graphic form and easily matches the pattern with condition of scene of social distance.

Table 1 Social distance measurement process

Decimal Degrees Places Distance 0 1 111 km 1 0.1 11.1 km 2 0.01 1.11 km 3 0.001 111 m 4 0.0001 11.1 m 5 0.00001 1.11 m 6 0.000001 11.1 cm 7 0.0000001 1.11 cm 8 0.00000001 1.11 mm 9 0.000000001 111 m 10 0.0000000001 11.1 m 11 0.00000000001 1.11 m 12 0.000000000001 111 nm

[0021]In Table 1 describe distance measurement process that includes Decimal Places, Degrees and Distance.

[0022]In Table 2 describe distance calculation process that calculated distance between LL(Latitude-Longitude) value of Scene of social distance (SOSD)and LL(Latitude Longitude) value of all mobile numbers, that are moving in every two-two minute.

Table 2 Provided by Telecom Service Provider of all Mobile Numbers

Time Mobile Latitude Longitude Number Initial time xxxx1 yyyy zzzz

[002311n Table 3 describe that all distance are arrange in ascending order thus it will give a fast analysis of making social distance clusters.

Table 3 Rearrange table and also calculated the distance with respect to Scene of social distance

Mobile Time Latitude Longitude Distance Number xxxx1 Initial yyyy zzzz --------- time

Table 4 Social Distance for People

Color Distance set Social Distance(At Least Cluster Accuracy set 6 Feet Distance) Cluster Green [75, 82, 95, [8, 10, 13, 11, 12] More than 6 Feet(Good 90, 93] Social Distance) Cluster2 Blue [70, 82, 92, [6, 8, 11, 9, 12] 6 Feet and More than 6 91, 93] Feet (Sufficient Social Distance) Cluster3 Red [94, 100, 55, [5, 6, 4, 6, 3] Less than 6 Feet(Not 100, 50] Sufficient Distance)

[0024]In Table 4 describe that all distance are with cluster, color, accuracy set, distance set, social distance (at least 6 feet distance) thus it will give a fast analysis of making social distance clusters.

[0025]The proposed design is made showing of Scene of social distance (SOSD) making a green cluster of social distances shown in Figure 2. Where Scene of social distance between peoples done. In figure 2 showing social distance accuracy of cluster (Green) that have good accuracy compare to cluster2 (Blue) and cluster3 (Red).

[0026]The proposed design is made showing of Scene of social distance (SOSD) making a blue cluster of social distances shown in Figure 3. Where Scene of social distance between peoples done. In figure 3 showing social distance accuracy of cluster2 (Blue) that have good accuracy compare to cluster3 (Red) but not have good accuracy compare to cluster (Green).

[0027]The proposed design is made showing of Scene of social distance (SOSD) making a red cluster of social distances shown in Figure 4. Where Scene of social distance between people done. In figure 4 showing social distance accuracy of cluster3 (Red) that have no good accuracy compare to cluster (Green) and cluster2 (Blue).

Claims (1)

1. CLAIMS OF THE INVENTION:

   We claim: 1. The SOSD comprising: (a) Social distance three clusters are known as inputs for process of Scene of social distance model; (b) Making three clusters cluster (Green), cluster2 (Blue), cluster3 (Red)for Scene of social distance model; (c) Making accuracy set for Scene of social distance model; (d) Making distance set for Scene of social distance model; (e) The designing of the process using cases social distance (At Least 6 Feet Distance) taking decision for Scene of social distance model; 2. The SOSD according to claim 1, wherein the input parameter to the proposed structure is cluster outcomes by social distance. The input parameter can be cluster outcomes by any available method such as social distance (At Least 6 Feet Distance) taking decision etc; however, the methodology should be adopted for cases. 3. The SOSD according to claim 1, along with a process in various available clusters parameters of respective process used to connect the scene of social distance model to other systems. 4. The SOSD according to claim 1 having higher gained as compared to general method thus, a COVID-19 volunteer is observed at all the process. 5. The SOSD according to claim 1, wherein the proposed scene of social distance model will always show better performance as compared to the general method irrespective of the method used in designing both the social distance, i.e. general social distance and proposed scene of social distance model.