An Approach for Finding Anthropogenic Prediction Using Novel

Cluster Analysis Technique over Exploratory Approach Based on

Accuracy

S. Sumanth and N. Deepa

Saveetha University, Chennai, India

Keywords: Artificial Intelligence, Civilization, Novel Cluster Analysis Technique, Human Intervention, Machine

Learning, Resilience.

Abstract: The objective of this research is to make a specialty of anthropogenic prediction by way of using a better

cluster evaluation approach on a dataset by means of evaluating with the E exploratory Approach. Materials

and using an exploratory approach method with a sample size of 27 and a novel cluster analysis technique

with a sample size of 27, respectively, and a G power of 80%, an accurate anthropogenic prediction was made.

Results: The accuracy of the Innovative Cluster Analysis Method is 77.26%, which is somewhat more than

the accuracy of the Exploratory Approach, which is 73.07%. With p=0.294 (p>0.05), it demonstrates there is

no statistically significant difference comparing New Cluster Analysis Method as well as Exploratory

technique. Conclusion: When compared to the subsets of variables in the marginal distribution and the

exploratory approach, which has an accuracy of 73.07%, novel cluster analysis technique has a greater

accuracy of 77.26%.

1 INTRODUCTION

A fundamental aspect of artificial intelligence is

machine learning, which equips systems with the

capability to perform tasks and execute functions

without explicit programming. This facet of AI

focuses on the utilization of computers, as highlighted

by Hottung and Tierney (2022), to access data and

execute tasks autonomously. Much like the human

mind accrues expertise and knowledge, machine

learning, as elucidated by Rouhanizadeh,

Kermanshachi, and Nipa (2020), relies on input such

as educational records and knowledge graphs to

comprehend entities, domains, and the relationships

between them. As entities define civilization, the

application of deep learning can contribute to

enhancing urbanization.

The machine learning process initiates with

observations and resilience from challenges,

encompassing examples, direct experiences, or

training. It seeks patterns within data to subsequently

make inferences based on the provided examples. The

primary objective of machine learning, as emphasized

by Kopec, Shetty, and Pileggi (2014) and G.

Ramkumar et al. (2022), is to empower computers to

autonomously learn without human intervention,

enabling them to adapt and perform tasks

independently. This autonomy is especially beneficial

for computers in executing programs without

requiring human involvement, as noted by Qochuk,

Sharma, and Chatterjee (2020) and Padma, S et al.

(2022).

Numerous research papers in the field of machine

learning are frequently published in IEEE and

Science Direct. The IEEE Xplore digital library, for

instance, houses 5877 journals, while Science Direct

boasts 3455 articles. Additional citations are available

in 6203 papers from Springer and 7990 articles from

Google Scholar.

The issue of environmental pollutants is closely

tied to urbanization and commercial trends across the

entire spectrum of human intervention. Air pollutants,

highlighted by Sinnott and Guan (2018), are

identified as primary concerns in metropolitan

regions globally, exacerbated by increased

urbanization and societal development. Notably,

Tehran, the capital of Iran, grapples with air pollution

challenges, affecting both the well-being of its

residents and the city's resilience (Sinnott and Guan,

2018). Delavar et al. (2019) emphasize that a

significant portion of Tehran's air pollutants is

attributed to PM10 and PM2.5 pollution.

Sumanth, S. and Deepa, N.

An Approach for Finding Anthropogenic Prediction Using Novel Cluster Analysis Technique over Exploratory Approach Based on Accuracy.

DOI: 10.5220/0012567000003739

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Artiﬁcial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics (AI4IoT 2023), pages 96-102

ISBN: 978-989-758-661-3

Metallic performance, as studied by Mele and

Magazzino (2020), holds a prominent position as one

of the most widely used resources in human

intervention. This prominence traces back to the

industrial revolution, coinciding with the advent of

machine learning techniques and the establishment of

large iron and steel production units (Nicodemi,

1994). The material's resilience activities are

facilitated by its high resistance to stress at elevated

temperatures, atmospheric and corrosive agents, as

well as its ductility and ability to undergo plastic

deformation. This adaptability aligns with the

technological advancements in civilization.

In the realm of cluster analysis strategies, existing

conventional algorithms exhibit low precision and

sensitivity. Cluster analysis, a statistical method

organizing similar items into respective categories,

serves as a fundamental approach in human

intervention (Zheng, 2022). This method aims to

group devices based on chosen traits and attributes of

civilization. For exploratory analysis, it is imperative

that data sets do not contain redundancies, missing

values, or null values (Cipresso et al., 2018). The

primary objectives of exploratory analysis include

identifying faulty data points to facilitate their

removal, thus ensuring data cleanliness. Moreover, it

aids in understanding the relationships between

variables, offering a comprehensive perspective on

the data.

The Novel Cluster Analysis Technique, in

comparison to the exploratory approach, specifically

targets identifying air pollution levels in the

atmosphere, contributing to a more refined

understanding of environmental conditions.

2 MATERIALS AND METHODS

The study takes place in the data analytics lab of the

SIMATS School of Engineering, which is well-

equipped with advanced facilities for both output

production and research purposes. Two assessments

of assembly numbers were conducted with a sample

size of 27 (Williams et al., 2019). The calculations are

secured using 80% G power, a 95% confidence level,

an alpha value of 0.05, and a beta value of 0.2,

implemented through clinical software.

The dataset provides information on the condition

of contaminated states and their corresponding air

quality levels. Analyzing particulate matter levels in

the dataset simplifies the determination of air quality

(Borbet, Gladson, and Cromar, 2018).

Cluster evaluation is a method used in gadget

studying that attempts to locate clusters of

observations within a dataset. The intention of cluster

analysis is to find clusters such that the observations

within each cluster are quite similar to every different,

whilst observations in exclusive clusters are quite

specific from each other.

Table 1: Shows procedure for proposed algorithm.

Input: Assign the clusters present in the data set.

Output: Accuracy of the data set's existing clusters.

Step 1: First need to define some clusters and assign

variable k.

Step 2: Choose at random a few clusters for

computation from those defining clusters.

Step 3: After computing the clusters, assign data points

to each cluster.

Step 4: Now define squared difference between

information factors of the clusters

Step 5: Now assign statistical variables to each cluster

present in the data.

Step 6: Find mean of all clusters present in the data set

after computing the data set.

Table 2: Exploratory Approach is coded as procedure.

Input: Give insights to the data set

Output: Accuracy of the optimal factors present in the

data set.

Step 1: Maximum insights should be obtained from the

data set.

Step 2: Uncover the unlayered structure present in the

data set.

Step 3: Important variables present in the data set need

to be extracted.

Step 4: Find out if any outliner should be identified

inside the data set.

Step 5: Underlying assumptions should be tested in this

stage.

Step 6: Finally have to determine the optimal factors

present in the data set.

An Approach for Finding Anthropogenic Prediction Using Novel Cluster Analysis Technique over Exploratory Approach Based on

Accuracy

Table 3: Displays the accuracy raw data table for both the

exploratory approach and the novel cluster analysis

technique.

Sl.no

Novel Cluster Analysis

Technique

Accuracy (%)

Exploratory

Approach

Accuracy (%)

Table 4: Displays the N (27), Mean (73.07%),

Std.Deviation (13.862), Std.Error Mean values (2.668) and

N (27), Mean (74.26%), Std.Deviation (15.096), Std.Error

Mean values (2.905) for Innovative Cluster Analysis

Method.

Algorithm

Mean

Std.Deviation

Std.Error

Mean

Accuracy

Novel

Cluster

Analysis

Technique

77.26

15.096

2.905

Exploratory

Approach

73.07

13.862

2.668

Google Colab, akin to the Jupyter notebook,

offers a convenient platform for program execution.

Its browser-based interface allows users to write and

execute arbitrary Python code, making it particularly

useful for data analysis. This feature facilitates code

sharing, enabling others to easily run the provided

code snippets (Borbet, Gladson, and Cromar, 2018).

Novel Cluster Analysis Technique

Group 1 in sample preparation utilizes an innovative

cluster analysis technique, treating a cluster of

statistical objects as a singular group. Before

assigning labels to organizations in cluster analysis, it

is imperative to initially divide the data collection into

groups based on information similarity for

robustness. The primary advantage of category-based

clustering lies in its adaptability, enabling the

differentiation of valuable services that distinguish

various businesses.

Clustering analysis finds extensive applications in

various fields, including market research, resilience,

pattern recognition, data analysis, and image

processing. Additionally, it assists entrepreneurs in

identifying human intervention and understanding the

distinct groups within their customer base due to

increased urbanization. Entrepreneurs can effectively

categorize customer groups based on their purchasing

habits. In the realm of biology, clustering analysis

proves useful for classifying genes with similar

functionalities, providing insights into population

structures. It also aids in deriving taxonomies for

plants and animals through artificial intelligence. In

databases containing Earth statements, clustering

facilitates the discovery of areas with similar land

uses. Employing machine learning, it contributes to

the recognition of property groups in a city based on

housing type, cost, and geographic location.

Table 1 outlines the Novel Cluster Analysis

Technique procedure.

2.1 Exploratory Analysis

Exploratory analysis is sample preparation group 2

and analyzes data with the help of visual techniques,

through this it can discover patterns or check

assumptions with the help of graphical representation

and statistical representation. Exploratory analysis'

principal goal is to examine the data before adopting

any civilizational assumptions. It can aid in locating

glaring faults, in addition to better understanding data

trends, locating outliers, and locating intriguing

relationships between the variables. Exploratory

analysis consists of data of the improved

urbanization. It is divided into three parts: data

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

summarization, data visualization, and data

normalization is a part of artificial intelligence. The

data summarization is when summarizing the essence

of the particular dataset using certain key statistical

measures such as mean and standard deviation of the

human intervention. Data visualization as the name

suggests is going to be visually exploring the data by

plotting various types of graphs such as histograms

and bar plots. The data normalization is to adjust the

scales of the data. Table 2 represents the procedure

for exploratory analysis.

2.2 Statistical Analysis

The following statistics were calculated using IBM

SPSS version 26.0: standard deviation mean, standard

error mean, mean difference, significance, as well as

F value. The amount of particle matter and the air

quality are independent variables. Deforestation is a

dependent variable.et al. 2021).

3 RESULTS

The procedure for the New Cluster Analysis Method

is displayed in Table 1. The number of clusters that

are present in the data collection must first be

specified. After discovering the data set, it is

important to concentrate on computing the clusters

and determining the average of all the clusters as a

sign of some pointers to the data set. This ought to aid

in determining correctness. present in the data set that

contains this data. In order to locate the common

points across all the clusters, the dataset is first

partitioned into two sets of models. These models are

then assigned to various functions in order to

calculate accuracy.

The procedure for the exploratory approach is

shown in Figure 2. Get the data first, then look for any

interesting findings. Discover the data set's unlayered

structure after data collection. The data for the

unlayered structure should be shown in two sets. This

should be used to validate the described structures

and determine accuracy.

Table 3 displays the accuracy raw data table for

both the exploratory approach and the novel cluster

analysis technique.

Table 4 displays the N (27), Mean (73.07%),

Std.Deviation (13.862), Std.Error Mean values

(2.668) and N (27), Mean (74.26%), Std.Deviation

(15.096), Std.Error Mean values (2.905) for

Innovative Cluster Analysis Method.

Table 5 represents the statistical independent

sample T-Test values. Mean difference for the New

Cluster Analysis Method and Exploratory Approach

is 4.185, Standard Error difference is 3.944, and 95%

Confidence Interval, respectively. With p=0.294

(p>0.05), it demonstrates that there is no statistically

significant difference between the New Cluster

Analysis and Exploratory method.

By looking at Figure 1, you can compare the mean

accuracy of two algorithms—the exploratory

approach and the novel cluster analysis technique.

Using a bar graph, the accuracy of both lost and

gained value is determined. When compared to the

exploratory approach, which has an accuracy of

73.07%, the Innovative Cluster Analysis Technique

has an accuracy of 77.26%.

4 DISCUSSION

The relevance of the air quality stages is determined

based on the results obtained using the enhanced

Cluster evaluation approach. The significance rate is

calculated entirely based on the outcomes attained

using unbiased factors. The accuracy of 77.26% is

exceeded by the significance value of 0.520 (p>0.05)

so there is no significance difference between them.

The accuracy of the New Cluster Analysis Method is

better than 73.07% for the exploratory approach. The

basic concept of anthropogenic prediction is to find

the quality of the air. This is the method to analyze

the data (Rouhanizadeh, Kermanshachi, and Nipa

2020) with the help of visual techniques. unfavorable

herbal disasters, have an effect on nations round the

world. To recover from a catastrophe, the managers

need to make diverse decisions using artificial

intelligence; however, what makes the submit-

disaster selection precise is lack of time to make the

optimal civilization decision. A complete expertise of

obstacles to recuperation can result in development of

rules that help prevent delays in healing technique

and therefore (Rouhanizadeh and Kermanshachi

2021; M R et al. 2019) effects in resiliency. Although

numerous studies had been achieved to become aware

of the healing obstacles, they did no longer offer a

complete overview of limitations and their

classification based on their techniques and also by

the improved urbanization. This paper aimed to

identify as well as categorize boundaries to effective

submit-disaster recuperation with a focal point on

hurricanes. The (M R et al. 2019) Researchers,

scientists, and educators are becoming more

interested in plant-based fibers that have been

extracted because of their application in polymer

composites, as well as their long-term viability in

machine learning and environmental friendliness.

An Approach for Finding Anthropogenic Prediction Using Novel Cluster Analysis Technique over Exploratory Approach Based on

Accuracy

Table 5: Shows the statistical independent sample T-Test values. Mean difference for the New Cluster Analysis Method and

Exploratory Approach is 4.185, Standard Error difference is 3.944, and 95% Confidence Interval, respectively. With p=0.294

(p>0.05), it demonstrates that there is no statistically significant difference between the New Cluster Analysis as well as

Exploratory method.

sig

Sig

(2-tailed)

Mean

difference

Std.Error

Difference

Lower

Upper

Accuracy

Equal

variables

assumed

.41

.52

1.06

.29

4.18

3.94

-3.73

12.10

Equal

variables not

assumed

1.06

51.62

.29

4.18

3.94

-3.73

12.10

Figure 1: Comparison among New Cluster Analysis Method (77.26%) as well as Explorative Approach (73.07%) depending

on accuracy. New Cluster Analysis Method has a higher average accuracy than Exploratory Approach. New Cluster Analysis

Method vs Exploratory Approach on X axis, Mean Accuracy on Y axis. +/-2 SE Error Bar.

The learning and extraction of the fibers affect the

characteristics of herbal fibers. Mechanical retting,

dew retting, or water retting techniques are used to

remove natural plant fibers. Mainly (Vardhan,

Kumar, and Panda 2019) For life to exist on earth,

water is an incredibly important supply. The

overpopulation, human activities, rapid

industrialisation, unskilled exploitation of natural

water resources, and unplanned urbanization have all

had a negative impact on the water quality. According

to Kaushik et al. (2022), "heavy metals" are a class of

metals and metalloids with atomic densities greater

than 4000 kg/m3. At their extremely low

concentrations, heavy metals are poisonous by nature

and can seriously harm both humans and animals.

This is relevant to artificial intelligence. Runoff from

agriculture and industrial discharges are two ways

that these heavy metals get into the aquatic system,

according to machine learning. The use of antibiotics

to treat bacterial infections has long been a tenet of

contemporary medicine (Kraemer, Ramachandran,

and Perron 2019). Antibiotic abuse and widespread

overuse, however, have resulted in unexpected

consequences that call for significant adjustments in

policy to be mitigated. Two major categories of

antibiotic misuse and overuse corollaries are

discussed in this review. The factors affecting the

anthropogenic conditions are a main issue of

anthropogenic climate change is international

warming, which refers to a sluggish warming of the

earth caused by an unnatural (human-precipitated)

growth of the greenhouse impact by the help of the

artificial intelligence, as concentrations of

greenhouse gasses growth generally from the burning

of fossil fuels (coal, oil, and herbal fuel). The

limitations of anthropogenic prediction are

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

100

deforestation, rising of earth atmosphere based on

machine learning. The future work of anthropogenic

prediction is used to find out the polluted air present

in the atmosphere.

5 CONCLUSION

Anthropogenic prediction makes determining the

amount of air pollution in the atmosphere easier. The

accuracy rate of the Novel Cluster Analysis

Technique (77.26%) is significantly higher than the

Exploratory Approach (73.07%) while using the

Novel Cluster Analysis Technique and Exploratory

Approach data sets, as well as when comparing the

two techniques.

REFERENCES

Borbet, Timothy C., Laura A. Gladson, and Kevin R.

Cromar. (2018). “Assessing Air Quality Index

Awareness and Use in Mexico City.” BMC Public

Health 18 (1): 538.

Cipresso, Pietro, Irene Alice Chicchi Giglioli, Mariano

Alcañiz Raya, and Giuseppe Riva. (2018). “The Past,

Present, and Future of Virtual and Augmented Reality

Research: A Network and Cluster Analysis of the

Literature.” Frontiers in Psychology 9 (November):

2086.

Delavar, Mahmoud, Amin Gholami, Gholam Shiran,

Yousef Rashidi, Gholam Nakhaeizadeh, Kurt Fedra,

and Smaeil Hatefi Afshar. (2019). “A Novel Method for

Improving Air Pollution Prediction Based on Machine

Learning Approaches: A Case Study Applied to the

Capital City of Tehran.” ISPRS International Journal

of Geo-Information.

https://doi.org/10.3390/ijgi8020099.

G. Ramkumar, G. Anitha, P. Nirmala, S. Ramesh and M.

Management Principle using Intelligent Wavelet

Transformation based Water marking Scheme,"

International Conference on Advances in Computing,

Communication and Applied Informatics (ACCAI),

Chennai, India, 2022, pp. 1-7, doi:

10.1109/ACCAI53970.2022.9752516.

Hottung, André, and Kevin Tierney. (2022). “Neural Large

Neighborhood Search for Routing Problems.” Artificial

Intelligence.

https://doi.org/10.1016/j.artint.2022.103786.

Kaushik, Pallavi, Renu Khandelwal, Neha Rawat, and

Mukesh Kumar Sharma. (2022). “Environmental

Hazards of Heavy Metal Pollution and Toxicity: A

Review.” FLORA AND FAUNA.

https://doi.org/10.33451/florafauna.v28i2pp315-327.

Kopec, Danny, Shweta Shetty, and Christopher Pileggi.

2014. Artificial Intelligence Problems and Their

Solutions. Mercury Learning and Information.

Kraemer, Susanne A., Arthi Ramachandran, and Gabriel G.

Perron. (2019). “Antibiotic Pollution in the

Environment: From Microbial Ecology to Public

Policy.” Microorganisms 7 (6).

https://doi.org/10.3390/microorganisms7060180.

Lavrentieva, Olena O., Ihor O. Arkhypov, Olexander I.

Kuchma, and Aleksandr D. Uchitel. (2020). “Use of

Simulators Together with Virtual and Augmented

Reality in the System of Welders’ Vocational Training:

Past, Present, and Future.”

https://doi.org/10.31812/123456789/3748.

Mele, Marco, and Cosimo Magazzino. (2020). “A Machine

Learning Analysis of the Relationship among Iron and

Steel Industries, Air Pollution, and Economic Growth

in China.” Journal of Cleaner Production.

https://doi.org/10.1016/j.jclepro.2020.123293.

M R, Sanjay, Suchart Siengchin, Jyotishkumar

Parameswaranpillai, Mohammad Jawaid, Catalin Iulian

Pruncu, and Anish Khan. (2019). “A Comprehensive

Review of Techniques for Natural Fibers as

Reinforcement in Composites: Preparation, Processing

and Characterization.” Carbohydrate Polymers 207

(March): 108–21.

Qochuk, Benjamin, Priyanshi Sharma, and Aditya

Chatterjee. 2020. Problems with AI (Artificial

Intelligence).

Padma, S., Vidhya Lakshmi, S., Prakash, R., Srividhya, S.,

Sivakumar, A. A., Divyah, N., ... & Saavedra Flores, E.

I. (2022). Simulation of land use/land cover dynamics

using Google Earth data and QGIS: a case study on

outer ring road, Southern India. Sustainability, 14(24),

16373

Rouhanizadeh, Behzad, and Sharareh Kermanshachi.

(2021). “Barriers to an Effective Post-Recovery

Process: A Comparative Analysis of the Public’s and

Experts' Perspectives.” International Journal of

Disaster Risk Reduction.

https://doi.org/10.1016/j.ijdrr.2021.102181.

Rouhanizadeh, Behzad, Sharareh Kermanshachi, and

Thahomina Jahan Nipa. (2020). “Exploratory Analysis

of Barriers to Effective Post-Disaster Recovery.”

International Journal of Disaster Risk Reduction.

https://doi.org/10.1016/j.ijdrr.2020.101735.

Sinnott, Richard O., and Ziyue Guan. (2018). “Prediction

of Air Pollution through Machine Learning Approaches

on the Cloud.” 2018 IEEE/ACM 5th International

Conference on Big Data Computing Applications and

Technologies (BDCAT).

https://doi.org/10.1109/bdcat.2018.00015.

Vardhan, Kilaru Harsha, Ponnusamy Senthil Kumar, and

Rames C. Panda. (2019). “A Review on Heavy Metal

Pollution, Toxicity and Remedial Measures: Current

Trends and Future Perspectives.” Journal of Molecular

Liquids. https://doi.org/10.1016/j.molliq.2019.111197.

Williams, A. Park, A. Park Williams, John T. Abatzoglou,

Alexander Gershunov, Janin Guzman‐Morales, Daniel

A. Bishop, Jennifer K. Balch, and Dennis P.

An Approach for Finding Anthropogenic Prediction Using Novel Cluster Analysis Technique over Exploratory Approach Based on

Accuracy

101

Lettenmaier. (2019). “Observed Impacts of

Anthropogenic Climate Change on Wildfire in

California.” Earth’s Future.

https://doi.org/10.1029/2019ef001210.

Zheng, Zhida. 2022. “A Novel Air Quality Prediction

Method Based on GAF and Dense Net.” (2022)

International Conference on Machine Learning and

Intelligent Systems Engineering (MLISE).

https://doi.org/10.1109/mlise57402.2022.00029.

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

102