Parential Chat Monitoring System for Child Abuse Eradication

DOI : 10.17577/IJERTCONV10IS08002

Download Full-Text PDF Cite this Publication

Text Only Version

Parential Chat Monitoring System for Child Abuse Eradication

Mr. A. Joshua Issac

Assistant professor

M.I.E.T. Engineering College Trichy, India

Nivetha Ravi

Computer Science and Engineering

M.I.E.T. Engineering College Trichy, India

Ishwarya Sasikumar

computer science and Engineering

M.I.E.T. Engineering College Trichy, India

Selva Prabha Ramachandran

computer science and Engineering

        1. Engineering college Trichy, India

          Abstract Girl Child safety is the problematic bone where numerous vituperative bedevilments take place in academies and sodalities. Sexual abuse of defendants/ malefactors, whether adult or juvenile, in community corrections, violates the law. It also violates their introductory mortal rights, impedes the liability of their successful re-entry into the community, and violates the Governments legal obligation to give safe and humane community corrections supervision. Therefore, in our proposed system, a machine literacy approach tends with Linear Supervised Machine Literacy and Random Matching to give out hands to help the parents to safeguard their children from this conditioning. Then parent monitoring system over the child exertion system is done with SMS, Exchange, and phone calls monitoring. Then trained database with multiple vituperative words keywords are get trained and they're matched with the childs SMS and exchanges. The terrain fully used textbook mining and pre- processing approach on a textbook bracket which classifies the textbook. The classified textbook is matched with the pre-trained database which will insinuate to the parent that the process of converse is vituperative. Therefore, the parent will get complete access to the pupil monitoring system. Suggestions grounded on this study are the need to develop psychoeducation for adolescents and families both as subjects and victims to avoid sexual importunity geste.

          Keywords Mlp, Nlp, Lsvm

          1. INTRODUCTION

            Over the once two decades Machine Literacy has come one of the reliances of information technology and with that, a rather central, albeit generally hidden, part of our life. With the ever-adding quantities of data getting available theres good reason to believe that smart data analysis will come indeed more pervasive as a necessary component for technological progress. The purpose of this chapter is to give the anthology an overview of the vast range of operations that have at their heart a machine literacy problem and to bring some degree of order to the zoo of problems. After that, we will badly some introductory tools from statistics and probability proposition since they form the language in which numerous machine literacy problems must be stated to come amenable to working Eventually, we will outline a set of fairly introductory yet effective algorithms to break an important problem, videlicet that of the bracket. More sophisticated tools, a discussion of further general

            problems, and a detailed analysis will follow after the corridor of the book. Text mining (also called textbook data mining or textbook analytics) is at its simplest, a system for drawing out content grounded on meaning and environment from a large body (or bodies) of the textbook or put another way, its a system for gathering structured information from unshaped textbooks. Its via textbook mining tools, for illustration, that numerous spam pollutants descry unwanted emails from your inbox, and how companies can anticipate, rather than simply reply to their client needs by sifting through millions of putatively unconnected data and discovering meaningful connections. Text mining also has a significant eventuality for academic operation and, at least when used in its introductory form, benefits from being a fairly straightforward and easy tool to master. Every time, millions of girls and boys around the world face sexual abuse and exploitation. Vituperative violence occurs everyplace in every country and across all parts of society. A child may be subordinated to abuse or exploitation at home, at academy or in their community. The wide use of digital technologies can also put children at threat. Most frequently, abuse occurs at the hands of someone a child knows and trusts. At least 120 million girls under the age of 20 about 1 in 10 have been forced to engage in illegal conditioning or perform other illegal acts, although the factual figure is probably much advanced. Roughly 90 percent of adolescent girls who report forced illegal conditioning say that their first perpetrator was someone they knew, generally a swain or a hubby. But numerous victims of vituperative violence, including millions of boys, no way tell anyone. Although vituperative violence occurs everyplace, pitfalls swell in exigency surrounds. During fortified conflict, natural disasters and other philanthropic extremities, women and children are especially vulnerable to vituperative violence including conflict- related vituperative violence, intimate mate violence and trafficking for illegal exploitation as well as other forms of gender- grounded violence. Vituperative violence results in severe physical, cerebral and social detriment. Victims witness an increased threat of social insulation and cerebral trauma. Some victims may resort to parlous actions like substance abuse to manage with trauma. And as child victims reach majority, vituperative violence can

            reduce their capability to watch for themselves and others. While vituperative violence is unnaturally a crime of power, it's decreasingly driven by profitable motives. The internet has opened a fleetly growing global request for the product, distribution and consumption of child illegal abuse accoutrements, similar as photos and vids. When online, children may be susceptible to illegal compulsion and in- contact illegal abuse by malefactors who essay to wring them for content and fiscal gain. The dangerous morals that immortalize vituperative violence take a heavy risk on families and communities too. Utmost children who face vituperative experience other kinds of violence. And as abuse and exploitation come settled, progress towards development and peace can stall

            with consequences for entire societies. UNICEF plays a crucial part in precluding and responding to vituperative violence worldwide both in exigency and non- emergency surrounds through programmers, hook-ups, and advocacy. Encyclopedically, we make advocacy tools and develop specialized guidance for violence forestallment and response, helping to insure services is applicable and sensitive to the requirements of survivors. We work nearly with mates on a variety of global enterprises, including the Global Partnership to End Violence against Children, and Together for Girls and we cover Global Alliance to End Child illegal Exploitation Online. At the public position, we work with governments to develop and strengthen laws and programs and to increase access to justice, health, education and social services that help the child and adolescent survivors recover. We also invest in public forestallment programmers to change social morals that blink vituperative violence and immortalize a culture of silence. Throughout all we do, we concentrate on supporting children and parents. We work directly with children to make their knowledge on how and where to seek help and protection; and with parents, preceptors and grown-ups to identity signs of abuse and make sure children admit ongoing care.


  1. Sexual offenses against children: socal learning theory and dark web reinforcement

    As cybercrime exertion decreasingly uses anonymous technologies, the dramatic growth of child sexual exploitations on the dark web has posed a challenge to law enforcement agencies. Data were gathered in 2019 and 2020 through face-to-face interviews with a chick in New Taipei City, Taiwan. This paper extends the operation of the social literacy proposition to online child sexual abuse geste. Research testing Akers's social literacy proposition has been confined to sexual offenses against children on the dark web. A social literacy conception of discriminational association, delineations, reproduction, and discriminational underpinning is illustrated and supported by online child sexual abuse geste findings. Before it occurs, precluding child sexual abuse has come a critical issue and necessary trouble from all areas of society family caring, academy

    instruction, community-grounded treatment, and social values. An innovative social literacy strategy to battle online child sexual abuse is proposed to reduce juvenile delinquency on the dark web.

  2. Analysis Of The Impact Of Abuse On Children By Using Big Data Method In Computer – A Case Study In Jakarta, Indonesia

    Child abuse can beget cerebral and Physical effect for children. Family factors and parenthood can be causes of child abuse. The end of the study was to determine the effect of child abuse in Indonesia by the big data system in computer. This Qualitative study was conducted on Jakarta. Jakarta is a sanctum that made by Jakarta government for child abuse victim and the specific position were being secret for victim safety). Study sample size was 50 children(victim of child abuse). Data was collected using 2 types, by interview and cerebral test. Out of 50 children there are 21 actors from sexual abuse, 16 actors from physical abuse, and 12 actors from cerebral abuse. From the interview 95 percent of the children having a cerebral trauma and posttraumatic stress complaint (PTSD) and from the cerebral test prove that posttraumatic stress complaint( PTSD) in children were veritably high. IQ and EQ are dwindling than the normal bone. The effect of child abuse in Indonesia veritably high and every time the cases increase. It suggested for parents to know further parenthood style and avoid child abuse to reduce the side effect of child abuse.

  3. Ethical Issues Of Child Abuse By Parents And Other People In Makassar City, Indonesia

Children are members of the community who are weak both physically and in the fulfillment of their right to the determined age of their right occasionally divided inversely into their parents. Acts of violence against children are acts of abuse that are carried out by parents or others toward children. Forms of violence that do can be physical, cerebral, sexual, to social. Physical violence in Indonesia is now one of the pitfalls for children and is known as a retired ménage tragedy. In the meantime, nearly always the crime that occurs in children in the family, by society in general, isn't seen as a crime. In this study, the system uses a quantitative approach, by using figures to describe characteristics, videlicet the ethical problem of child abuse by parents and others. With a study sample of 10 people and calculating the average also assessed using the Independent Sample T- test the average value of abuse by parents is more dominant than abuse by other people. It's because parents allowed it was a way of educating that the child would come a chastened.

D. Pornography And Child Sexual Abuse Detection In Image And Video: A Comparative


With the growing quantum of pornography content over Internet and cases of Child Coitus Abuse(CSA) material possession and distribution, there's a rising demand for automatic discovery of similar content especially in certain surroundings similar as educational or work places. The donation of this paper is three-fold. First, we

present a critical review of automatic pornography and CSA discovery in images and vids. Second, we give an empirical evaluation of five named pornography discovery approaches representing traditional skin discovery grounded as well as more recent deep literacy grounded styles. The evaluations are performed under common criteria using two intimately available pornographic databases. Eventually, we assess these styles on a dataset of real- world CSA material handed by Spanish Police Forces. This study observes that for pornography or CSA discovery, the styles involving multiple features perform better than those using simple features like skin color or a single image descriptor. It also plants that deep literacy grounded styles outperform all of the other styles and report current state-of-the-art.

E. Intelligent Interpretation And Analysis Of Child Sexual Abuse Forensic Evidence: A Preliminary Study

In sexual abuse case, the victim's body is the most important source of physical substantiation. Meanwhile, medical help plays a part as part of the police disquisition in forensic examination. Forensic substantiation in the sexual assault will be collected because of the intimate nature of this substantiation and different special chops are demanded to conduct a detailed examination. Thus, in child sexual abuse cases, misgivings regarding of colorful environment in decision timber can lead to failure of successful disquisition. Colorful types of substantiation data need to be anatomized and interpreted to come out with a dependable and precise report. Therefore, this paper presents a primary study of this content in order to gain a result of decision making in the child sexual abuse forensic field through an intelligent decision support system.

III. Existing system

Earlier the past month, 43 girl children have been engaged in bad conversation with their school men staff has been found out by the parent and they are alleged from the school by the paper. Recently a 2 girl committed suicide because of the smart phone call and message threatening from their staff. These problems prevailing in the existing system where there is no solution to safe guard this girl child from these unknown sexual traitors. Normal existing system gives an identification of these sexual harassments in social media where they will be based on the comments and tweets placed. The normal image based analysis is placed with speak back language which done on disclosure language classification. The learning layers are get analyzed and known well at the level of the pattern found. In the existing system, usually, text classification system is used on spammer detection. Social media tag-based sexual harassment identification is done using the normal speak-back language system. Girl child suffers a lot in school, working place, and college because of many sexual traitors. Image analysis of the activitys detection are done in the existing process.



    In our proposed system, easy monitors the activities of the girl child moving. The monitoring mainly focuses on sexual harassment and threatening the girl child. These monitoring mainly gives the classification of the sexual harassments based chats which will be monitored and made intimation to the parent. Then the parent will monitor the chats, SMS, Phone calls and the location of the girl using the Global Positioning System (GPS). The chats will be continuously intimated to the parent side where they will completely know the chats and calls take by them. This classification will be done with fuzzy based String searching and random field classification system. Here the literature survey produces each concept where each one has an issue. To maintain an overall safety monitoring android based access system is used. The identification of the girls safety is proposed in our system which saves the child from these sexual offenders. The system creates a safety environment where the girls can be monitored anywhere they go.In our proposed a system, easy monitors the activities of the girl child moving. The monitoring mainly focuses on the sexual harassment and threatening over the girl child. These monitoring mainly gives the classification of the sexual harassments based chats which will be monitored and made intimation to the parent. Then the parent will monitor the chats, SMS, Phone calls and the location of the girl using the Global Positioning System (GPS).

      1. Suspicious word collection

        Extraction of the abusive words is the problematic one. The identification of the intrinsic private chat collection of data from the internet is the issue one where the harassment keywords on all language with normal English typing have been collected in the dataset. The new dataset creation should be based on the linguistic markers with the corpus of word using the known suspicious forum of words which will be the anonymized data. The abusive words are downloaded from the UCI machine learning.

      2. Keyword training

        The dataset training will be the one where the admin need to train the dataset. The dataset training will be the creation of a remote server with dataset access. Self disclosure annotation model will be generated with the server side database storage system.Three parts classification system using the NLP on universal model final tuning has been proposed. The hyper parameters are drop with multiple tuning levels pattern. The weight matrix will be managed on the recurrent connective patterns.

      3. Preprocessing

        The pre-processing will focus on noise reduction, white space elimination, special character elimination, etc. Then the system substantially gives on some NLP processing system which will be given on the converse dispatches system.

      4. Intimation and Monitoring

    After the identification of the harassment chat, the user profile will be intimated to the parent. The parent will completely monitor the girl childs harassment profile with the GPS coordinates of the latitude and longitude too. With the system, they will completely know their childs activities of location, SMS, Chats, and phone calls management. Thus, the parents can safeguard by taking the correct action against the harassment traitor and the girl child.


    The developed machine learning-based child abuse safety environment software helps the parent to monitor their child. The system will help with the elimination of child harassment and reduce suicide cases in India. Natural Language Processing plays a vital role in the assignment of the linguistic language system. The training and the matching of the string will be done easily with the given fuzzy-based string matching and random field algorithm. Supervised language-based text classification has been done. This knows the harassment text classification over multiple profiles and makes intimation to the parent. In future enhancement systems, the identification of the illegal harassers that can be directly identified by the police will be made. The intimation of the IP with MAC address, latitude, and longitude of the person can be retrieved for unknown id threatening.


[1] Chi-cheng Yang; En-tien Ma; Da-Yu Kao, "Sexual Offenses against Children: Social Learning Theory and Dark Web Reinforcement", International Conference on Advanced Communication Technology (ICACT), 2021

[2] Yao Yu; Weiwei Liu, "Ethical Issues of Child Abuse-A Cultural Comparison", International Conference on Public Health and Data Science (ICPHDS), 2020

[3] Jie Ren, "Analysis of the Impact of Abuse on Children by Using Big Data Method in Computer – A Case Study in Jakarta, Indonesia", International Conference on Artificial Intelligence and Education (ICAIE), 2020

[4] Yao Yu; Weiwei Liu, "Ethical Issues of Child Abuse by Parents and Other People in Makassar City, Indonesia",

International Conference on Public Health and Data Science (ICPHDS), 2020

[5] Yao Yu; Weiwei Liu, "Ethical Issues of Child Abuse by Parents and Other People in Makassar City, Indonesia", International Conference on Public Health and Data Science (ICPHDS), 2020

[6] Andrews.C, Fichet.E, Ding.Y, Spiro E.S, and Starbird .K, (2016)Keeping up with the tweet- dashians: The impact ofofficialaccounts on online rumoring, in Proceedings of the 19th ACM Conference on Computer-Supported Cooperative Work & Social Computing. ACM, pp. 452465.

[7] Castillo.C, Mendoza.M, and Poblete.B, (2011)Information credibility on twitter, in Proceedings of the 20th international conference on World wide web. ACM, pp. 675684.

[8] Chapelle .O, Schlkopf.B, and Zien. (2010) A, Semi-Supervised Learning, 1st ed. The MIT Press

[9] Ferreira.W and Vlachos.A, (2016 )Emergent: a novel data-set for stance classification, in Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: Human language technologies, pp. 11631168.

[10] Giasemidis.G, Singleton.C, Agrafiotis.I, Nurse J.R.C, Pilgrim

.A, Willis.C, and Greetham D.V,( 2016) Determining the veracity of rumours on twitter, in Social Informatics: 8th International Conference, SocInfo 2016, Bellevue, WA, USA, November 11-14, 2016, Proceedings, Part I, E. Spiro and Y.-

Y. Ahn, Eds. Cham: Springer International Publishing, pp. 185 205.

[11] Gupta.A, Lamba.H, Kumaraguru.P, and Joshi.A, (2013) Faking sandy: characterizing and identifying fake images on twitter during hurricane sandy, in Proceedings of the 22nd international conference on World Wide Web. ACM, pp. 729 736

[12] Hamidian.S and Diab M.T, (2015 )Rumor detection and classification for twitter data, in Proceedings of the Fifth International Conference on Social Media Technologies, Communication, and Informatics (SOTICS), pp. 7177.

[13] Jin .Z, Cao.J, Zhang.Y, and Luo.J,( 2016) News verification by exploiting conflicting social viewpoints in microblogs. in AAAI, pp. 29722978.

[14] Suresh Babu V, Anu George, Image Forgery Detection using Global, Local and Histogram Features, Technology and Future Journal of Science and Technology, Volume 1, Issue 1, page no.52-58,2014.

[15] S Vijayan and Jestin V K Suresh Babu V, Survey On Identification Of Microaneurysm In Retinal Image Using Support Vector Machine International Journal of Applied Engineering Research, volume 10, Issue 67, Page no 17-22, 2015.