CC BY
30
№ 2 (116)
февраль, 2024 г.
NEURAL NETWORK NOWADAYS: HOW TO "TWEE" ENGLISH SPEAKING SKILLS
IN SECONDARY SCHOOL
Sofia Ushkova
Master's student, The Moscow City University, Russia, Moscow E-mail: sofia. ushkova2534@yandex. ru
НЕЙРОСЕТЬ В НАШИ ДНИ: КАК «ПОДТЯНУТЬ» НАВЫКИ РАЗГОВОРНОГО АНГЛИЙСКОГО В ОСНОВНОЙ ШКОЛЕ
Ушкова София Станиславовна
магистрант
Московского городского педагогического университета,
РФ, г. Москва E-mail: sofia.ushkova2534@yandex.ru
ABSTRACT
The paper is devoted to the study of the Twee neural network in English lessons. The paper puts forward arguments for the efficiency of the neural network as a tool for teaching speaking. Therefore, it exemplifies the exercises for the communicative skills development in secondary school. The results obtained confirm the validity of using the Twee neural network in English lessons. The study provides teachers with a new effective exercise system of teaching speaking to teenagers.
АННОТАЦИЯ
Статья посвящена изучению нейросети Twee на уроках английского языка. В статье приводятся аргументы в пользу эффективности нейросети как средства для обучения говорению. В связи с этим рассматриваются примеры упражнений для развития коммуникативных умений в основной школе. Полученные результаты подтверждают эффективность использования нейросети Twee на уроках английского языка. Исследование предоставляет учителям новую результативную систему упражнений для обучения говорению детей подросткового возраста.
Keywords: the Twee neural network, communicative skills, secondary school, learning tool, exercise set.
Ключевые слова: нейросеть Twee, коммуникативные умения, основная школа, средство обучения, комплекс упражнений.
Neural networks are now emerging as a powerful tool for improving oral proficiency in secondary school. Language learning, particularly oral language learning, has long been a challenge due to the limited variety of dialogues and communicative scenarios presented in traditional textbooks. By bridging the gap between traditional language learning and modern technological solutions, it is necessary to consider neural networks as an alternative learning tool that can be used alongside, rather than instead of, traditional ones. In this way, the learning process can be diversified with modern learning tools that are appealing to learners.
With an emphasis on key research initiatives, the paper provides a brief overview of innovative studies of P.V. Sysoev, M.G. Evdokimova, Z.S. Kurbanova, N.P. Ismailova, A.S. Nesterov, G.D. Svirina that use neural networks to enhance language learners' oral proficiency. P.V. Sysoev notes that chatbots, as programmes created on the basis of machine learning and natural language, allow to organize foreign language speech practice of secondary school and university students and thus contribute to the further formation of their foreign language communicative competence [8, p. 203].
Moreover, neural networks can act as an interlocutor, engaging the learner in a discussion. In this way, it is possible to practice a foreign language outside the classroom [1, p. 177].
The demands of the modern world have raised a considerable interest in exploring new potentials and opportunities resulting from the digitalization of education. One such prospect is related to the integration of artificial intelligence in the form of neural networks-software algorithms that function on the basis of enumerating values to solve problems [6, p. 297].
Researchers have studied various aspects of neural network application. P.V. Sysoev highlighted the possibilities of chatbots and voice assistants as programmes based on artificial intelligence that facilitate interaction with humans to solve communicative tasks [8]. M.G. Evdokimova considered the definition of the concept of 'artificial intelligence', the history of its development, application and operation principles in education [1].
Neural network technologies prove to be useful in assessing students' communication skills. Through neural networks, one can build an individual learning trajectory
Библиографическое описание: Ushkova S. NEURAL NETWORK NOWADAYS: HOW TO 'TWEE' ENGLISH SPEAKING SKILLS IN SECONDARY SCHOOL // Universum: психология и образование : электрон. научн. журн. 2024. 2(116). URL: https://7universum. com/ru/psy/archive/item/16768
№ 2 (116)
and assess the training efficiency by comparing the results before and after. Timely identification of knowledge gaps can be addressed through personalized learning scenarios [2, p. 151].
Applying next-generation neural networks and intelligent didactic tools allows the creation of personalized learning paths that promote the development of personality-oriented education and improve the quality of student learning [4, p. 310].
In late 2022, the Twee service came out, operating based on ChatGPT principles to convert each neural network request into exercises and tasks for English learning.
Let us look at the features of the Twee neural network for teaching speaking in secondary school.
1) Text generation in a specified language on various topics with customizable narrative modes, vocabulary, and grammatical forms. Using a given text makes it possible to organize a discussion between learners. Moreover, generating texts can be used to compare textbooks or personal texts, allowing students to assess their writing strengths and weaknesses.
2) Dialogue generation. Neural networks are capable of creating dialogue on demand. It is also possible to set the topic and background of a communicative situation. These dialogues can be used as an illustration of certain communicative situations and as a model for students to make up their own dialogues.
3) Generation of new versions of previously composed text. A neural network can generate a simplified or complicated version of the original text. This will allow the same text to be adapted according to the learner's level of knowledge.
4) Question generation. The neural network can generate questions both on a specific topic and on the basis of a given text. In addition, the service makes it possible to generate not only the question itself, but also answer variants to it [5, p. 34].
5) Quote generation. The neural network is able to offer a number of quotes when the prominent person's name is given. This exercise boosts the discussion between learners.
6) Generation of communicative situations with the requested vocabulary. In this way, students learn new words in context.
Thus, the Twee neural network is able to generate tasks that test language literacy and text comprehension.
Nevertheless, The Twee neural network has got shortcomings.
1) The difficulty of usage. The Twee is an artificial intelligence-based platform that requires electronic devices and an Internet connection. People who do not have stable access to technology can face technical challenges.
2) Limitations in social communication. Although the Twee offers the opportunity to communicate in English through AI, it cannot completely replace live communication, which limits the development of social skills.
3) The lack of complete personalized feedback. It may reduce the efficiency of Twee learning: teaching English using artificial intelligence with the generation of questions and assignments [7].
However, they do not prevent learners from achieving the planned learning outcomes.
февраль, 2024 г.
According to its characteristics and structure, the Twee is an AI-based system to create didactic materials for learning English. Applying them, the teacher will be able to organize students' work with reading texts at a new level, with relatively little time, and create a basic dictionary for students, as well as other tools to develop written and oral speech, improving grammatical and listening skills.
The Federal educational standard of secondary general education includes certain personal, meta-subject and subject results for students to achieve. Let us consider the system of exercises generated by the Twee neural network and reveal their linguodidactic potential in terms of achieving the required learning outcomes.
The exercise set includes the following steps:
1) Sample dialogue generation.
At this stage students engage in comprehensible reading, analysis of the communication situation, and the dialogue type. Besides, students examine the emotional background of the dialogue. The stage is completed by role-playing the dialogue and working with vocabulary by means of contextual guessing, where cases of using new lexemes are discussed.
2) Question generation for the dialogue.
The stage includes checking the understanding of the dialogue content and the semantic structure of the text.
3) Communication situation generation.
Students practice lexical units in different communicative contexts. In the end, students act out the situations and highlight grammatical structures to be used in their own dialogues.
4) Composition and presentation of students' own dialogues relying on new vocabulary.
Students act out their own dialogues in front of the class using the sample dialogue under new communication conditions.
5) Reflection of the activity, correction of dialogues.
Students analyze their work: communicative purpose
achievement, emotional colouring, application of new lexical units, grammatical structures, mistakes, task complexity. Students discuss how to improve their performance.
Let us consider a concrete example of the exercises generated by the Twee neural network based on the Starlight 6 teaching package, module 5 'Helping Hands'.
The lexical units to compose the sample dialogue are as follows: drought, flood, earthquake, earthquake, forest fire, tornado, tsunami, hurricane, natural disaster, endangered animals, pollution, global warming [1].
Grammar: The Present Simple Tense.
Communicative task: ask your partner about natural disasters.
Type of dialogue: an inquiry dialogue.
Level: A1-A2.
1) Generating a dialogue with new vocabulary.
Sophie: Hi, John. Do you know about bad things that can happen to nature?
John: Yes, I do.
Sophie: Can you tell me more?
John: Well, sometimes there's no rain for a long time and it makes everything very dry. That's called a drought. It can be hard to find water.
№ 2 (116)
Sophie: Oh, that's not good. What else can happen?
John: There can be too much rain and the water covers everything. That's a flood. And there are big shaking movements in the ground called earthquakes.
Sophie: Scary! Are there any other kinds of bad things like this?
John: Yes, there are also fires in the forest, spinning winds called tornadoes, giant waves called tsunamis, and huge storms called hurricanes.
Sophie: Wow, so many! Have you ever seen one of these things?
John: Luckily, no. But my town had a tornado before.
Sophie: That sounds really scary. What happens to animals when these things happen?
John: They can lose their homes or get hurt. Some animals might even become endangered because of these bad things.
Sophie: That is sad. Is there anything we can do to help stop these bad things from happening?
John: Many smart people think pollution and global warming might make them worse. We have to take care of our world!
Sophie: Right. Thanks for telling me all of this, John.
John: No problem, Sophie.
2) Dialogue question generation.
Questions:
1. What is it called when there's no rain for a long time and everything becomes very dry?
2. What happens when there is too much rain and the water covers everything?
3. What are the big shaking movements in the ground called?
4. What are fires in the forest called?
5. What are spinning winds called?
6. What are giant waves called?
7. What are huge storms called?
Correct answers:
1. It is called a drought.
2. When there is too much rain and the water covers everything, it is called a flood.
3. The big shaking movements in the ground are called earthquakes.
4. Fires in the forest are called wildfires.
5. Spinning winds are called tornadoes.
6. Giant waves are called tsunamis.
7. Huge storms are called hurricanes.
3) Generation of communicative situations with new vocabulary.
1. The news just reported a drought in our area. We need to conserve water.
I agree. Let's fix any leaks and take shorter showers.
2. Did you see the damage from the flood? It's devastating.
Yes, it's so sad. Let's donate to the relief efforts.
3. I heard there was an earthquake in California. I hope everyone is safe.
Me too. I hope the building codes were strong enough to withstand it.
4. My friend's house was destroyed in the earthquake.
That's terrible. I hope they have a place to stay.
5. The forest fire is spreading quickly. We need to evacuate.
февраль, 2024 г.
I'll grab our important documents and some clothes. Let's go.
6. There's a tornado warning. We need to take shelter immediately.
I'll grab the emergency kit. Let's go to the basement.
7. Did you hear about the tsunami in Japan? It's devastating.
Yes, I saw the pictures. It's heartbreaking.
8. The hurricane is expected to hit our town tomorrow.
We should stock up on supplies and stay indoors.
9. Natural disasters can be so unpredictable.
Yes, we should always be prepared for the worst.
10. Many animals are becoming endangered due to human actions.
We need to raise awareness and protect their habitats.
11. Pollution is harming our environment. We need to reduce our carbon footprint.
I agree. Let's start by using reusable bags and recycling.
12. Global warming is a serious issue. We need to take action now.
Yes, we should reduce our energy consumption and support eco-friendly initiatives.
4) Making up students' own dialogues based on the sample.
5) Reflexion. At this stage, students self-evaluate their activities and results.
Hence, the exercise set provides a variety of contexts for the lexical units in various communicative situations. This contributes to more effective memorization of new lexical items, as well as to the development of speaking through constant oral interaction between students. As a result of regular practice on different topics, students will learn new lexemes, basic grammatical constructions and, in particular, will be prepared to communicate together in a foreign language.
The exercise set facilitates the achievement of personal, subject and meta-subject students' results. Let us examine them.
The personal outcomes include the awareness of one's own and others' emotions and ability to regulate one's own emotional state; readiness for a variety of collaborative activities, commitment to mutual understanding and mutual assistance; self-analysis skills, acceptance of the right to make mistakes for oneself and others; adaptability to stresses and changing social and information conditions; acceptance of oneself and others without judgement; stress recognition by students, enabling them to adjust decisions and actions; active participation in solving practical tasks, ability to initiate, plan and independently fulfil objectives.
Subject outcomes involve the communication skills: engaging in courteous dialogues, prompting action-oriented discussions, and utilizing questioning dialogues with up to five responses from each participant; the comprehension-oriented reading: independently reading and understanding authentic texts adapted for the comprehension of unfamiliar words; the lexical dimension of speech: defining and employing lexical units in oral and written communication that cater to specific communicative situations, ensuring logicality and coherence
№ 2 (116)
• 7universum.com
UNIVERSUM:
, nCMXO/IOrHfl H 05PA30BAHHE
foeBpaflb, 2024 r.
in expressions; the grammatical dimension of speech: identifying and utilizing learned morphological forms and syntactic structures in both oral and written English communication; the compensatory skills: applying contextual inference during reading and incorporating key words, phrases, and a plan to formulate personal statements.
Meta-subject outcomes cover the cognitive learning strategies: utilize questions as an exploratory tool to understand; select, analyze, systematize, and interpret information of various types and forms of representation; the regulative learning behaviour: demonstrate self-control and reflection; make adjustments to activities; regulate emotional expressions; consciously engage with others and their opinions; acknowledge the right to make mistakes for oneself and others; the communicative learning skills: perceive and articulate judgments; express emotions in line with communication conditions; comprehend the intentions of others; demonstrate respect for the interlocutor; pose questions during dialogue or discussion; plan the organization of collaborative efforts [3].
In conclusion, the neural network application emerges as a highly effective tool to enhance spoken language skills in secondary school English lessons. The demonstrated efficacy of exercises generated by the Twee neural network underscores its ability to facilitate students meeting the specified requirements outlined in the Federal basic educational programme. Moreover, the exercise set's adaptability stands out, allowing for customization based on individual student needs, considering the diverse range of task complexities and accommodating variances in learning styles. The adaptability not only ensures a tailored learning experience for each student but also signifies the potential for continued optimization and improving teaching methodology in response to evolving educational requirements. Thus, the integration of neural networks in language learning not only proves benefits to achieve curriculum objectives but also highlights the dynamic and responsive nature of modern educational technologies in fostering effective language acquisition.
References:
1. Evdokimova M.G., Agmaliev R.T. Linguodidactic potential of artificial intelligence systems / M.G. Evdokimova, R.T. Agmaliev. — Economic and Social Research. — 2023. — № 2 (38). — P. 173—191. [in Russian].
2. English language. 6th grade: educational. for general education institutions and schools with depth studying English language / [TO. M. Baranova, D. Dooley, V.V. Kopylova and others]. — M.: Express Publishing: Education, 2013. — 184 pp. — (Starlight). — ISBN 978-5-09-027652-8 [in English].
3. Eltunova I.B., Nesterov A.S. Creation of accounts users of the domain of the local area network / I.B. Eltunova, A.S. Nesterov. — Modern teacher education. — 2021. — № 11. — P. 150—154. [in Russian].
4. Federal Educational Programme of Basic General Education : official site. URL: https://static.edsoo.ru/pro-jects/fop/index.html#/sections/200205 [in Russian].
5. Kurbanova Z.S., Ismailova N.P. Neural networks in the context of digitalization of education and science / Z.S. Kurbanova , N.P. Ismailova. — The world of science, culture, education. — 2023. — № 3 (100). — P. 309—311. [in Russian].
6. Kravtsova A.G. Chatgpt—3: perspectives of application to foreign language teaching / A.G. Kravtsova. — The world of science, culture, education. — 2023. — № 3 (100). — P. 33—35. [in Russian].
7. Svirina G.D., Shashok P.A. Application of neural network technology in teaching / G.D. Svirina, P.A. Shashok. — Global science. — 2018. — № 6 (15). — P. 297—304. [in Russian].
8. Sysoyev P.V., Filatov E.M. Method of the development of students' foreign language communication skills based on practice with a chatbot / P. V. Sysoyev, E. M. Filatov. — Perspectives on science and education. — 2023. — № 3 (63). — P. 201—218. [in Russian].
9. The Twee Neural Network : web—site.— 2023. URL: https://twee.com/ [in Russian].
