February 05, 2022

"Due to the pandemic, online jobs are increasing. However, some online jobs are fraudulent, stealing personal and vital information. These fake jobs can be detected and classified using advanced deep learning and machine learning algorithms."

The growth of the internet has made recruitment much more accessible. Moreover, the current pandemic has influenced the current trend in job recruitment. Online recruitment has increased candidate access and streamlined processes, bridging the gap between recruiters and candidates. Candidates can now apply to many jobs online, based on their specialisation. Moreover, online recruitment helps users find more jobs and hire the best candidates. This recruitment process allows the recruiters to reach qualified candidates worldwide.

Additionally, Online recruitment ensures that the client hires the best candidate. In recent times, Candidates have been screened using Facebook and LinkedIn. Pre-employment screening, personality assessment, and testing allow companies to select qualified candidates and thus improve efficiency. Human intervention is minimal in this process.

However, some of these job postings are just fake jobs designed to snatch data from potential applicants. Hackers may hack a candidate's laptop or another electronic device to obtain sensitive information when they apply for these positions. Cybercriminals collect their victims' personal information and sell or use it for their gain on the dark web or even years later. Using the proper dataset, analysis, and cleaning, as well as appropriate machine learning and deep learning algorithm implementation, these fake jobs can be identified and data theft prevented.

According to several studies, the detection of review spam, According to several studies, detecting review spam, email spam, and fake news has garnered considerable attention in the field of online fraud detection.

Spam Detection:

Job seekers frequently use online forums to share their thoughts on their purchased products. It could serve as a guide for other customers to shop for their products. Spammers can profit from manipulating reviews in this context, so researchers must develop techniques for detecting spam reviews. We can accomplish this solution by using Natural Language Processing (NLP) to extract features from reviews.

Email Spam Detection:

Spam emails, including unwanted bulk mails, are standard in user mailboxes. Gmail, Yahoo Mail, and Outlook service providers have Spam filters using Neural Networks to combat this issue. For example, for content-based filtering, email spam detection is used as well as heuristics, memory or instance filtering, and adaptive approaches.

Fake News Detection:

Twitter's echo chamber effect and malicious user accounts are the hallmarks of fake news on the social media platform.

Moreover, There are Numerous works on detecting fake jobs and related issues such as seeing fake news and spam emails. Most of these papers employ machine learning algorithms to see fictitious jobs within a pool of legitimate employment. Most frequently used methodologies in various research papers for detecting fake jobs include the following:

• Detecting fake jobs by utilising a publicly available dataset.
• Machine learning algorithms to classify fake and legitimate job postings from a pool of job postings.
• You can improve machine learning algorithms by using ensemble classification modelling.
• Data is subjected to semantic analysis to bridge relationships and detect fake jobs and false news.
• It is detecting and classifying categorical data with N-gram analysis.
• Extracting features from textual data can be done using semantic and natural language processing techniques.
• It is possible to achieve higher classification accuracy by employing complex algorithms like XGBoost.

The researchers say that supervised learning algorithms are the first step in identifying job posting scams. The machine learning approach employs several classification algorithms for detecting fake posts. A classification tool detects and alerts the user to fake job postings in this case.

Conclusion

Online recruitment saves money on communication. However, spotting employment scams is an essential step in ensuring that job seekers only receive genuine offers of employment. The researchers propose several machine learning algorithms to deal with the detection of employment scams. This problem continues to be addressed by a growing number of researchers. The research community will soon provide us with an improved solution.

Dr Nivash Jeevanandam PhD,
Researcher | Senior Technology Journalist