Detection of Careless Responses in Online Surveys

Aim:

          The main aim of this project is to provide Online Survey Using Answering Behavior on Smart Phone.

Abstract:

       Some respondents make careless responses due to the ‘‘satisficing,’’ which is an attempt to complete a questionnaire as quickly and easily as possible. To obtain results that reflect a fact, detecting satisficing and excluding the responses with satisficing from the analysis targets are required. One of the devised methods detects satisficing by adding questions that check violations of instructions and inconsistencies. However, this approach may cause respondents to lose their motivation and prompt them to satisficing. Additionally, a deep learning model that automatically answers these questions was reported. This threatens the reliability of the conventional method. To detect careless responses without inserting such screening questions, machine learning (ML) detection using data obtained from answer results was attempted in a previous study, with a detection rate of 55.6%, which is not sufficient from the viewpoint of practicality. Therefore, we hypothesized that a supervised ML model with a higher detection rate could be constructed by using on-screen answering behavior as features. However, (1) no existing questionnaire system can record on-screen answering behavior and (2) even if the answering behavior can be recorded, it is unclear which answering behavior features are associated with satisficing.

EXISTING SYSTEM

          To detect careless responses with a higher detection rate, we hypothesized that answering behavior on a touch-screen can be used. However, no current questionnaire system records on-screen answering behavior and even if answering behavior is recorded, the features of answering behavior associated with satisficing are unclear. Hence, we designed
and developed a plug-in that can record answering behavior such as scrolling, choosing and changing options, and entering and deleting text (hereafter referred to as ‘‘Operation
Logger’’). Operation Logger was developed as a plug-in that runs on Lime-Survey, a widely used online questionnaire system. Therefore, the questionnaire system can be used in
the same way as without the plug-in, and the answering behavior can be recorded by simply adding Operation Logger to the survey. Thus, respondents do not need to install any additional applications or perform any other additional operations.

Problem Definition:     

• Getting sensor-data from mobile devices will not be accurate for classification in most times.
• Based on the physical usage most of the user mentality cannot be predicted.
• When we implement the monitoring technique in client-side app it results in performance issue due to numerous services need to be initiated at background.

PROPOSED SYSTEM

            We propose a novel approach to detect the seriousness or rate of carelessness of a person who is undergoing an online survey application. We can implement sentiment analysis to detect the state of responses. We can integrate machine learning techniques with some python libraries namely open-cv, eyeball tracking. This will give satisfying results. When the user undergoing survey their face video will be captured and file will be sent to python server at the end of the survey session. After processing the video frames it will respond the rate of carelessness. Based on this approach we can develop a high performing responsive client app for taking survey.

Advantages:

• High performance client app
• Machine Learning classification is implemented with python library.