Design of Flight Cadet Emotion Intelligent Prediction System

Based on Mobile Phone Software Data

Hao Wen

, Zhiyong Zhang

, Congcong Lv

, Xiaolong Feng

and Guan Yuan

National University of Defense Technology, Changsha, 410000, China

The People’s Liberation Army Hong Kong Garrison, Hong Kong, 999077, China

Keywords: Big Data Technology, Flight Cadet, Emotion Prediction.

Abstract: With the further development of big data technology, the use of big data technology to collect the data of

mobile phone software, applied to the analysis of flying cadets’ mood, can help flight cadets, flight coaches

and flight doctors to understand the emotional state of flight cadets, find the signs of bad emotions in time,

help them get rid of bad emotions, and have a good effect on improving cadets’ learning enthusiasm and

mental health. In this paper, through the literature method, analysis method, system method and other

methods, the relevant mobile phone software data, as well as more mature and practical hardware and

software, are reasonably selected to design the emotional intelligent prediction system. The system describes

how to design the emotional intelligent prediction system according to the principle of emotion prediction,

the design of the intelligent prediction system, and the application prospect of the intelligent prediction

system, etc. We also provide a theoretical basis and application scope for the next step of developing the

emotional intelligent prediction system.

1 INTRODUCTION

In his congratulatory letter to the Fifth World Internet

Conference in 2018, President Xi Jinping stressed

that “today’s world is experiencing a larger and

deeper scientific and technological revolution and

industrial transformation. Modern information

technologies such as the Internet, big data, and

artificial intelligence have continuously made

breakthroughs, the digital economy has flourished,

and the interests of all countries have become more

closely linked. To add new features to the

development of the world economy, we urgently

need to accelerate the development of the digital

economy.” Flight cadets have great learning pressure,

long training time, difficult operating subjects, high

intensity of physical load and other characteristics,

which makes flight cadets have higher requirements

for emotional management and control than ordinary

college cadets. Big data technology is a product of the

deep development of scientific and technological

revolution and industrial transformation.

Accelerating the application of big data technology is

inevitable to keep up with the development of The

Times, as well as the inevitable acceleration of

military development. As high school cadets just

entering college, they need an adaptation process.

During this period, if you can better understand the

emotional state of the flight cadets, you will better

help the cadets to control their emotions. Instructors,

aviation doctors and others can also timely enlighten

the bad emotions of flight cadets, which is helpful for

improving the enthusiasm of flight cadets to learn and

train and maintaining mental health. This not only

conforms to the law of the development of the times,

but also meets the requirements of military

informatization construction.

2 THE PRINCIPLE OF

SENTIMENT PREDICTION

FOR MOBILE PHONE

SOFTWARE DATA

With the development and popularization of mobile

phone 4G technology, mobile phone users have

become more and more easy to obtain Internet

information through mobile phones, and people are

more and more willing to use mobile phone software

to obtain text, pictures, music, videos, and other

information on the Internet. A lot of data is generated

Wen, H., Zhang, Z., Lv, C., Feng, X. and Yuan, G.

Design of Flight Cadet Emotion Intelligent Prediction System Based on Mobile Phone Software Data.

DOI: 10.5220/0012041100003620

In Proceedings of the 4th International Conference on Economic Management and Model Engineering (ICEMME 2022), pages 635-640

ISBN: 978-989-758-636-1

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

635

during the use of the mobile phone software. Through

a series of data processing and analysis, it helps to

predict the emotional state of mobile phone users.

2.1 Classification of Mobile Phone

Software Data

We divide mobile phone software data into mobile

phone software data, user interaction data and user

personal data. Mobile phone software data mainly

refers to the data provided to us by the mobile phone

software itself, including text, picture, audio, video

and other four forms. The software data refers to the

data provided by the software company; User

interaction data refers to data such as text, picture,

audio and video generated by users in the process of

using software. For example, user comments,

uploaded photos, music, video, etc., are open and

interactive to a certain extent. It is important to note

that this kind of data is publicly available to other

users, which is different from personal data. User

personal data refers to some non-public data

generated by users during the use of software, which

can be divided into expressive data and recorded data.

Expressive data mainly refers to the data that users

convey their ideas in the form of text, picture, voice,

and other forms in software. These data are public

and private. Recorded data mainly refers to some

track-based data generated by users in the process of

using software, including tour records, use time,

voice, expression, etc.

2.2 Mobile Phone Software Data Input

We set up MySQL relational database management

system on the computer. The SQL language used by

MySQL is the most common standardized language

for accessing databases. The data storage is divided

into basic data and cadet data. The basic data includes

mobile phone software data and user interaction data,

while the cadet data refers to the user’s personal data.

2.2.1 Input of Mobile Phone Software Data.

Mobile phone software data is divided into four

forms of text, pictures, music, and videos, which we

process separately. For text data, we directly store it

in .txt format to MySQL. For image data, we convert

the image into binary data stream and save it to

MySQL. For music data, we store music text

information, including song name, singer, lyrics,

music introduction, music style and other

information, into MySQL as text data. For video

data, we divide it into video text information and

video voice information. Text information, including

video name, video author, video introduction, video

style, and text information in a video, is stored to

MySQL as text data. For the voice information in the

video, we use speech recognition software to convert

it into text information, and then store it in MySQL

as text data.

2.2.2 Input of User Interaction Data.

There are four forms of user interaction data, namely

text, picture, music, and video, which are divided into

user interaction comment data and user interaction

sharing data. User interaction comment data is

divided into comment text, comment emoticons,

comment voice, comment emoticons pictures and

other data. Comment text is stored in TXT format to

MySQL. The comment memes of the mobile phone

software are all corresponding text to summarize the

meaning of the memes, and we store the

corresponding text of each expression in TXT format

to MySQL. Convert comment speech into text-based

data and store it in TXT format to MySQL. Because

the resolution of emoticon pictures is low, the content

is more complex, the total amount is small and

concentrated in WeChat and QQ chat software, and

the current picture recognition technology is difficult

to convert it into text-based data, we do not enter such

data. Interactive shared data mainly refers to the text,

picture, music, video, and other data uploaded by

users on mobile software platforms, which can be

viewed and forwarded by other users.

2.2.3 User’s Personal Data Input.

We divide the user’s personal data into expressive

data and input data. For expressive data input, we

process the user’s data in the form of text, picture,

voice and other forms in the software according to the

above input method of mobile phone software data

and store it in the cadet database of MySQL software.

As for the input of recorded data, we convert the tour

records, usage time records, voice and facial

expressions generated by users in the process of using

the software into words in TXT format and store them

in the cadet database of MySQL software.

2.3 Analysis of Mobile Phone Software

Data

After we input the data of mobile phone software into

the database, we need to analyse the data to get the

mood prediction. We divide the database into the

basic database and the cadets’ database.

ICEMME 2022 - The International Conference on Economic Management and Model Engineering

636

2.3.1 Selection of Emotion Judgement

Vocabulary.

We selected eight basic emotions (joy, trust, fear,

surprise, sadness, disgust, anger, and hope) and

feedback emotions from the psychologist Plutchik’s

emotional wheel model, in which feedback emotions

are an extension of one of the basic emotions. We use

eight basic emotions as the basic emotion vocabulary.

After classifying 500 common emotion words on the

Internet as appropriate feedback emotions, feedback

emotion words with feedback emotion meaning are

formed.

2.3.2 Basic Database Data Processing.

We tagged the number of basic sentiment words and

feedback emotion words for text, images, music, and

videos. According to the proportion of basic

emotional vocabulary accounting for 70% and

feedback emotional vocabulary accounting for 30%,

a value is calculated, and then the basic emotion of

the text, pictures, music, and videos is judged and

marked. For example, a music file contains 30 basic

emotion words for “joy”, 70 feedback emotion words

for “joy” and 100 feedback emotion words for

“surprise”. According to the calculation, the weight

number of “joy” emotion words is

30*70%+70*30%=42, and the weight number of

“surprise” emotion words is 100*30%=30. Because

42 > 30, we judged the music to be emotionally

happy, labelled as the basic emotion of “joy”.

2.3.3 Data Processing in Cadets’ Database.

Expressive data is treated the same as a basic

database. After marking relevant basic emotional

words on the text, picture, music and video of the

cadet's tour, the proportion of expressive data of the

cadet was calculated according to the number of basic

emotional words and the total number of all

emotional words. For example, the total number of

expressive data emotion words of a cadet in one day

is 100, among which the basic emotion word “joy” is

20 and the basic emotion word “fear” is 10, then the

proportion of expressive data of the cadet “joy” is 0.2

and the proportion of expressive data of the cadet

“fear” is 0.1.

Visit records and usage time records in recorded

data are marked according to the basic mood words

corresponding to the text, picture, music and video in

the basic database. For example, cadet A watched

“joy” video for 15 minutes and “surprise” article for

10 minutes today. Then the proportion of recorded

data of the cadet was calculated according to the

proportion of the duration of each basic emotion to

the total duration of the tour. For example, take one

day as a measurement unit, cadet A watch “joy”

articles for 20 minutes, listen to “sad” music for 30

minutes, “joy” videos for 60 minutes, “fear” pictures

for 10 minutes. The duration of the tour is 20+60=80

minutes for the joy class, 30 minutes for the sadness

class, 10 minutes for the disgust class, and the total

duration of the tour is 120 minutes. The proportion of

recorded data of cadet’s “joy” was 80/120=0.66, the

proportion of recorded data of “sand” was

30/120=0.2; and the proportion of “hate” recorded

data was 10/120=0.08.

2.4 Prediction of Flight Cadets'

Emotions

After processing the data of the two databases, we

select the data in a period of time to calculate the

proportion of expressive data and recording data of

cadets' basic emotions, and then calculate the

proportion of expressive data *60%+ recording data

*40% of the same basic emotions, so as to obtain the

predicted value of each basic emotion. The eight

values are sorted to find out which basic emotions the

cadet has more during this period. At the same time,

the heart rate, pressure, blood pressure and other

physiological indicators of the traineess wearable

devices during this period can also assist in the

judgment of basic emotions, and finally predict the

emotional state of the flight trainees during this

period.

3 DESIGN OF STRUCTURE AND

FUNCTION OF FLIGHT

CADETS’ EMOTION

INTELLIGENT PREDICTION

SYSTEM

After analysing the principle and combining the

existing hardware, software and technical conditions,

we designed the emotion intelligent prediction

system as shown in figure 1:

Design of Flight Cadet Emotion Intelligent Prediction System Based on Mobile Phone Software Data

637

Figure 1: Emotional intelligence prediction system.

3.1 Data Collection Subsystem

The data collection subsystem consists of hardware

and software. The hardware includes the mobile

phones of flight trainees and smart monitoring

bracelets, and the software includes information

mobile software, social mobile software, shopping

mobile software, music mobile software, video

mobile software and so on. Flight cadets use mobile

software through mobile phones and monitor

physiological index data through wristbands. The

hardware is collecting data as well as generating it.

This data is obtained by negotiating with the handset

software vendor for an API. This subsystem is mainly

the media of data generated by the daily use of mobile

phone software by flight trainees, and the original

data we need to collect.

3.2 Data Processing Subsystem

The data processing subsystem consists of hardware

and software, including a hardware laptop computer,

a mouse and MySQL software. We use MySQL

software to build basic database and user database.

Then the data obtained from the API port of mobile

phone software and smart wristband are input into the

basic database and user database according to their

respective categories. Then, the contents of the two

basic databases are retrieved to match the emotion

words with the data, so as to prepare for the next step

of analysis. The subsystem mainly injects the data

generated in the daily use of mobile phone software

into two databases for certain processing.

3.3 Emotion Analysis Subsystem

We compared the data in the basic database and user

database of MySQL software, mainly quantifying the

emotion of each user within a certain period

according to our weight formula. Quantified results

were used to predict the mood of flight cadets. The

subsystem mainly calculates the data generated by

the flight trainees' daily use of mobile phone software

according to a certain weight and predicts the

emotional state of flight trainees within a certain

period.

3.4 Emotion Display Subsystem

The subsystem of emotion display consists of

hardware and software, including a monitor, a mobile

phone and supporting software. By designing

supporting software, we can call the data stored in

MySQL software in real time and display the results

of mood analysis subsystem through HDMI cable

connected to the computer of mood analysis

subsystem for mood prediction results. The computer

of mood analysis subsystem transfers data to

supporting software. The software will transmit the

results of intelligent prediction to the mobile phone

through the Internet for real-time display, the results

of intelligent prediction will be conducive to the

ICEMME 2022 - The International Conference on Economic Management and Model Engineering

638

flight cadets and other relevant personnel to

understand their emotional state for reference.

4 APPLICATION PROSPECT OF

FLIGHT CADET EMOTION

INTELLIGENT PREDICTION

SYSTEM

4.1 Help Flight Cadets Understand

Their Emotions Better

We know that flying is a high-risk job, especially for

flight cadets, only by having a clear understanding of

themselves and a good state can they better learn to

fly and operate flight. Our usual cognition of our

emotional state is sometimes inaccurate and too

subjective, and there is a lack of quantification of

relevant data. The emotion intelligence prediction

system can help flight cadets have a quantitative

understanding of their own emotions. Through the

supporting software of the emotional intelligent

prediction system, cadets can observe some of their

own data, so as to better understand their emotions

and quickly adjust their emotional state, which will

greatly improve the safety of flight.

4.2 Help Flight Instructor to

Understand Trainees’ Emotional

State

Flight teaching can achieve better teaching efficiency

under the condition that flight cadets have a good

emotional state and a high degree of concentration.

To improve the efficiency of flight instruction and

better complete the teaching, flight instructors often

need to understand the emotional state of flight

cadets. Flight cadets with negative emotional states

are prone to distraction, inattention, and decreased

observation during flight, which makes learning less

efficient. Through the emotion intelligent prediction

system, the flight instructor understands the

emotional state of the flight cadets before the flight

teaching, which is conducive to adjusting the learning

content, optimizing the learning method, and

improving the learning efficiency.

4.3 Help flight Doctors to Understand

the Emotional State of Trainees

The flight cadets’ operation of the aircraft is not only

related to their physiological state, but also to their

psychological state. To ensure the safety of the flight,

the flight doctors will often test the psychological

state of the flight cadets before the flight. In the past,

they all judged the status of flight cadets through

psychological scales, psychological interviews, etc.

Through the more accurate and objective emotional

intelligent prediction system for detection, it can

greatly improve the efficiency of the flight doctors’

detection of the psychological state of the flight

cadets and improve the safety of the flight.

5 CONCLUSION

The rapid development of big data technology and

artificial intelligence technology shows that it is

feasible to apply its ideas and methods to the

psychological training of flight cadets and conform to

the law of development. In order to better combine

the big data technology and artificial intelligence

technology with the characteristics of the flight

cadets, we design the emotion intelligent prediction

system, which has a certain theoretical significance.

In this paper, according to the principle of emotion

prediction, the design of intelligent prediction

system, the application prospect of intelligent

prediction system, etc., we describe how to design

emotion intelligence prediction system, laying a solid

theoretical foundation for the next practical

application. Although the emotion intelligent

prediction system is designed in this paper, the

research is still in the theoretical stage and is still in

the preliminary research stage. We have not yet

applied validation of the intelligent emotion

prediction system, nor have we dialectically

examined the relationship between the subsystems. In

the next step, we will conduct environmental

construction and application verification of

intelligent prediction systems and strengthen the

dialectical dialectic of the relationship between

various subsystems of environmental construction.

REFERENCES

Hui Yi, F.: Research on emotion recognition based on heart

rate variability. Biomedical Engineering Research 39

(2), 128-132 (2020).

Lin Mao, F.: Application of emotion Wheel in information

technology education and teaching. New Curriculum

30 (12), 142 (2018).

Ling Zhang, F.: Design of distance teaching system based

on artificial intelligence network. Modern Electronic

Technique 44 (2), 131-134 (2021).

Design of Flight Cadet Emotion Intelligent Prediction System Based on Mobile Phone Software Data

639

Minjing Wang, F.: Dissemination features and mining

system design of news big data based on artificial

intelligence. Manufacturing Automation43 (7), 91-95

(2021).

Qishi Hu, F.: Design of multimedia database online

integration system based on artificial intelligence.

Modern Electronic Technique 44(2), 127-130 (2021).

Ran Li, F.: A review of text Emotion Analysis. Journal of

Computer Research and Development 55 (1), 30-52

(2018).1

Ying Zhang, F.: Sentiment prediction of multi-source News

comments based on bidirectional hierarchical Semantic

Model. Journal of Computer Research and

Development 55 (5), 933-944 (2018).

Yong Liu, F.: Target emotion prediction based on deep

learning. Application of Computer Systems 29 (6),

211-217 (2020).

ICEMME 2022 - The International Conference on Economic Management and Model Engineering

640