TR2023017507A2

TR2023017507A2 - AN IMPROVEMENT SYSTEM FOR EDUCATIONAL INSTITUTIONS

Info

Publication number: TR2023017507A2
Application number: TR2023/017507
Authority: TR
Inventors: Kalkan Berna
Original assignee: Turkcell Teknoloji̇ Araştirma Ve Geli̇şti̇rme Anoni̇m Şi̇rketi̇
Filing date: 2023-12-18
Publication date: 2024-02-21

Abstract

Bu buluş, alınan onaylar doğrultusunda öğrencilerle yapılan anketler ve görüntülü görüşmeler üzerinde duygu analizi yaparak eğitim kurumlarının başarı sıralamasını belirleyip aynı zamanda karşılaşabilecekleri potansiyel sorunların ve gelişim alanlarının tahminlemesini yapmak üzere geliştirilen bir sistem (1) ile ilgilidir.This invention relates to a system (1) developed to determine the success ranking of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also to predict the potential problems and development areas they may encounter.

Description

TARIFNAME EGITIM KURUMLARINA YÖNELIK BIR IYILESTIRME SISTEMI Teknik Alan Bu bulus, alinan onaylar dogrultusunda ögrencilerle yapilan anketler ve görüntülü görüsmeler üzerinde duygu analizi yaparak egitim kurumlarinin basari siralamasini belirleyip ayni zamanda karsilasabilecekleri potansiyel sorunlarin ve gelisim alanlarinin tahminlemesini yapmak üzere gelistirilen bir sistem ile ilgilidir. Önceki Teknik Günümüzde egitim alaninda yasanan sikintilar çogunlukla anketler yapilarak sonuçlarin ögretmenler, psikologlar, egitim danismanlari tarafindan tamamen manuel sekilde ve kisisel bilgiler çerçevesinde degerlendirmeler yapilarak düzeltilmeye çalisilmaktadir. Bu degerlendirmeler kisitli imkanlar sebebi ile çogunlukla genel anlamda bir degerlendirme olup özele indirgenerek yapilamamakta ve degisen jenerasyona uygun olamayabilmektedir. Teknikte duygu analizleri yapilarak degisik is kollarinda iyilestirmelere yönelik çalismalar mevcuttur. Ancak teknikte ulasilan çalismalar arasinda ögrenciler özelinde düzenli araliklarla anketler ve görüntülü görüsmeler yapip ögrencilerin duygu analizi yapilarak gerçek fikirlerini çikararak, sorunlari ve bu sorunlara yönelik çözümleri tahminleyen bir sistem bulunmamaktadir. Bu sebeple mevcut teknikte bulunan çalismalar ve eksiklikler göz önünde bulunduruldugunda alinan onaylar dogrultusunda ögrencilerle yapilan anketler ve görüntülü görüsmeler üzerinde duygu analizi yaparak egitim kurumlarinin basari siralamasini belirleyip ayni zamanda karsilasabilecekleri potansiyel sorunlarin ve gelisim alanlarinin tahminlemesini yapmak üzere gelistirilen bir sisteme ihtiyaç duyuldugu anlasilmaktadir. Teknigin bilinen durumunda yer alan arastirma sonucunda ulasilan Arastirma dokümaninda, sesten metine (speech-to-text) özelligi olan, video arama ve ses kaydi üzerinden dogal dil isleme yöntemleri ile duygu analizi yapabilen ve anketlerle desteklenen bir müsteri hizmetleri sisteminden bahsedilmektedir. Bulus gerçek zamanli yönetici yardimi ile müsteri iletisim hizmetini uygulamaya yönelik sistemler ve yöntemlerle ilgilidir. Bir süpervizör, bir müsteri iletisim hizmetinde birden fazla araciyi denetleyebilmektedir. Bir bilgi islem kaynagi hizmet saglayicisinin bir hizmeti, bir bilgi islem kaynagi hizmet saglayicisinin bir hizmetindeki çok sayida ses baglantisini izleyebilmekte, çok sayida ses verisi için transkriptler olusturabilmektedir, olusturmak için bir dizi dogal dil isleme (NLP) teknigi kullanarak transkriptleri analiz edebilmektedir. Meta veriler, transkriptleri en azindan meta verilere dayali olarak kategorilerle etiketlemekte, transkriptlere, meta verilere ve kategorilere dayali olarak çok sayida baglantinin en az bir kismi için bilgi üretin ve bilgileri aracilarin bir süpervizörüne saglamaktadir. Sistem, görüntülü ve sesli görüsmelerle veri toplamaktadir. Sistem, yapay zeka altyapili bir modül ile görüsmelerin transkriptini olusturabilmektedir. Ayni zamanda, dogal dil isleme yöntemlerine dayanarak duygu analizi yapabilmektedir. Bulusun Kisa Açiklamasi Bu bulusun amaci, alinan onaylar dogrultusunda ögrencilerle yapilan anketler ve görüntülü görüsmeler üzerinde duygu analizi yaparak egitim kurumlarinin basari siralamasini belirleyip ayni zamanda karsilasabilecekleri potansiyel sorunlarin ve gelisim alanlarinin tahminlemesini yapmak üzere gelistirilen bir sistem gerçeklestirmektir. Bulusun Ayrintili Açiklamasi Bu bulusun amacina ulasmak için gerçeklestirilen "Egitim Kurumlarina Yönelik Bir Iyilestirme Sistemi" ekli sekilde gösterilmis olup, bu sekil; Sekil 1 Bulus konusu sistemin sematik görünüsüdür. Sekilde yer alan parçalar tek tek numaralandirilmis olup, bu numaralarin karsiliklari asagida verilmistir: 1 . Sistem 2. Uygulama 3. Veri tabani 4. Sunucu Alinan onaylar dogrultusunda ögrencilerle yapilan anketler ve görüntülü görüsmeler üzerinde duygu analizi yaparak egitim kurumlarinin basari siralamasini belirleyip ayni zamanda karsilasabilecekleri potansiyel sorunlarin ve gelisim alanlarinin tahminlemesini yapmak üzere gelistirilen bir sistem (1); -bir elektronik cihaz üzerinde yürütülen, en azindan ögrencilerin egitim kurumlarina ve aldiklari egitime yönelik sorularin yer aldigi anketler içeren bir arayüz sunmak üzere yapilandirilan en az bir uygulama (2), - uygulama (2) üzerinden ögrencilere sunulan anket sorularini, ögrencilerin verdikleri yanitlari kayit altina almak üzere yapilandirilan en az bir veri tabani (3), -herhangi bir uzak iletisim protokolünü kullanarak uygulama (2) ile veri alisverisi gerçeklestirmek, uygulama (2) ile kurulan iletisimle ögrencilerin anket sorularina verdikleri cevaplari veri tabanina (3) kaydetmek, görüntülü görüsme ile ögrencinin sorulara verdigi cevaplari alarak metne dönüstürmek ve veri tabanina (3) kaydetmek, dis sunuculardan güncel veri sözlügü verilerini alarak veri tabanina (3) depolamak, veri tabaninda (3) depolanan tüm verileri alarak duygu analizleri yapmak ve veri tabanina (3) egitim kurumu bazinda karsilasilabilecek sorunlar ve iyilestirmeleri belirleyip kaydetmek ve kurumlarla paylasmak üzere yapilandirilan en az bir sunucu (4) içermektedir. Bulus konusu sistemde (1) yer alan uygulama (2), en azindan ögrencilerin kendilerine tanimli bir bilgi ile giris yaparak egitim sistemi, sosyal çevresi ve egitim kurumlari ile ilgili anket sorularini görüntülemesine, sorulara yanitlar vermesine olanak saglayan bir arayüz sunmak üzere yapilandirilmaktadir. Bulusun bir düzenlemesinde uygulama (2) ögrencilerin kisisel profillerini olusturmasina ve ek yorumlar yazmasina olanak tanimak üzere bir arayüz sunmak üzere yapilandirilmaktadir. Bulus konusu sistemde (1) yer alan veri tabani (3), uygulama (2) ile sunucu (4) arasindaki iletisim sayesinde sunucunun (4) uygulamadan (2) aldigi ögrencinin uygulama (2) üzerindeki sorulara verdigi yanitlari kayit altina almak üzere yapilandirilmaktadir. Veri tabani (3) sunucu (4) tarafindan dis sunuculardan alinan güncel veri sözlügü verilerini, sunucunun (4) anketlere verilen cevaplar üzerinde yaptigi analiz sonuçlarini, sunucu (4) tarafindan belirlenen önerileri kayit altina almak üzere yapilandirilmaktadir. Bulus konusu sistemde (1) yer alan sunucu (4), veri tabani (3) üzerindeki verilere erismek ve veri tabani (3) üzerine veri kaydetmek üzere yapilandirilmaktadir. Sunucu (4), düzenli olarak ögrencilerle yapilan görüsmenin ses kaydini alarak speech to text (STT) yöntemi ile sesten metine çevirmek ve metinleri veri tabanina (3) kaydetmek üzere yapilandirilmaktadir. Sunucu (4) ögrencilere belli zamanlarda belli konularda görüslerinin soruldugu, metin olarak bu görüslerin alindigi anketleri uygulama (2) üzerinden sunmak ve alinan cevaplari veri tabanina (3) kaydetmek üzere yapilandirilmaktadir. Sunucu (4) ögrencilerin sorunlari ve memnuniyetlerini ölçmek amaçli sorulari egitim kurumuna iletmek ve alinan cevaplari veri tabanina (3) kaydetmek üzere yapilandirilmaktadir. Sunucu (4), sözlük içerikleri ve puanlamalarin yer aldigi güncel veri sözlügü verilerini dis sunuculardan alarak veri tabanina (3) kaydetmek üzere yapilandirilmaktadir. Sunucu (4) veri tabanina (3) kaydedilen sözlük verileri, ögrencilerin anket cevaplari ve görüsme verilerine NLP (Natural Language Processing-Dogal Dil Isleme Yöntemi) uygulamak üzere yapilandirilmaktadir. Sunucu (4) NLP uygulanan verilere sentiment analysis (duygu analizi) yöntemi uygulamak ve elde edilen verileri veri tabanina (3) kaydetmek üzere yapilandirilmaktadir. Sunucu (4) tüm kurumlardan alinan veriler üzerinden denetimli makine ögrenmesi yapmak ve kurumlarda çikabilecek sorunlar ve memnuniyetler hakkinda önceden tahminleme yapmak üzere yapilandirilmaktadir. Sunucu (4) dil analizinin yapilacagi dil özelliklerinin göz önüne alinmasiyla hazirlanmis olan her kelime için nümerik olarak bir kodlama metodu kullanmak üzere yapilandirilmaktadir. Kelimenin cümlenin neresinde kullanildigi, kelimenin sifat mi, yüklem mi oldugu, devrik bir cümle içerisinde mi kullanildigi seklindeki durumlar nümerik kodlamada fark yaratmaktadir. Sunucu (4) ayni metin içerisinde veya ayni kisinin birden fazla metni içerisinde yapilan analizleri karsilastirarak kisinin duygu durumunun normali olup olmadigini veya bir T anindan sonra belirli bir duyguya geçis yaptigini tespit etmek üzere yapilandirilmaktadir. Bulusun sanayiye uygulanmasi Bulus konusu sistemde (1), kamu ya da özel egitim kurumlarinda ögrencilerin görüslerini toplamak için anketler düzenlenecektir. Düzenli araliklarla belirlenen her yas kategorisinden ayni sayida kiz ve erkek ögrenci belirlenerek görüntülü görüsmeler yapilacak ve hem sosyal ortam hem de aldiklari egitim konusunda memnuniyetleri ölçümlenecektir. Bu anketlerden ve görüsmelerden alinan bilgiler disinda bu bilgileri veren ögrencinin görüsme sirasindaki duygu durumu önem arz etmektedir. Bilgileri verirkenki samimiyeti, bilginin dogruluk derecesi aslinda hissettiklerinin tam tersini söylemesi durumlari söz konusu olabilir. Ögretmenlerinden ve egitim kurumunun yönetiminden çok memnun oldugunu söyleyen bir ögrenci aslinda tam ters duygular içerisinde olabilir. Bu, kisinin kurdugu cümledeki yüklemin yerinden, devrik cümle kurmasina, kisa cevaplarla geçistirmesinden veya gereginden fazla cevabi uzatmasindan anlasilabilmektedir. Sunucu (4) duygu analizi ile tüm bu verileri irdeleyip ögrencilerin neye karsi sorunlar yasadigi, nelerden mutlu oldugu, ne yapilirsa daha verimli sonuçlar elde edilecegi seklindeki tüm detaylar ortaya çikarilarak ve bu detaylar veri tabanina (3) sürekli olarak kaydedilip sürekli olarak üzerinde degerlendirmeler yapilarak kendini egiten ve gelistiren bir sistem (1) ortaya çikarilmaktadir. Bu sayede tüm yeni gelismelerle ve tüm çaglardaki ögrencilerin egilimleri ile egitilen bir sistem (1) yaratilarak egitim kurumlarina tavsiyelerde bulunulmaktadir. Bulus konusu sistem (1) alinan onaylar dogrultusunda ögrencilerle yapilan anketler ve görüntülü görüsmeler üzerinde duygu analizi yaparak egitim kurumlarinin basari siralamasini belirleyip ayni zamanda karsilasabilecekleri potansiyel sorunlarin ve gelisim alanlarinin tahminlemesini saglamaktadir. Bu temel kavramlar etrafinda, bulus konusu "Egitim Kurumlarina Yönelik Bir Iyilestirme Sistemi (1)" ile ilgili çok çesitli uygulamalarin gelistirilmesi mümkün olup, bulus burada açiklanan örneklerle sinirlandirilamaz, esas olarak istemlerde belirtildigi gibidir. TR TR DESCRIPTION AN IMPROVEMENT SYSTEM FOR EDUCATIONAL INSTITUTIONS Technical Area This invention is related to a system developed to determine the success ranking of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also to predict the potential problems and development areas they may encounter. Prior Art Today, the problems experienced in the field of education are mostly tried to be corrected by conducting surveys and evaluating the results completely manually and within the framework of personal information by teachers, psychologists and educational consultants. Due to limited resources, these evaluations are mostly general evaluations and cannot be reduced to specifics and may not be suitable for the changing generation. There are studies aimed at improvements in different lines of business by performing sentiment analysis in the technique. However, among the studies available in the technique, there is no system that predicts problems and solutions to these problems by conducting surveys and video calls on students at regular intervals, analyzing students' emotions and extracting their real opinions. For this reason, considering the studies and deficiencies in the current technique, it is understood that there is a need for a system developed to determine the success ranking of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also to predict the potential problems and development areas they may encounter. In the research document obtained as a result of the research on the known state of the technique, a customer service system that has a speech-to-text feature, can perform emotion analysis with natural language processing methods over video call and voice recording, and is supported by surveys is mentioned. The invention relates to systems and methods for implementing customer communication service with real-time manager assistance. A supervisor can supervise more than one agent in a customer communication service. A service of a computing resource service provider can monitor multiple audio connections within a service of a computing resource service provider can create transcripts for large amounts of audio data can analyze the transcripts using a set of natural language processing (NLP) techniques to create Metadata tags transcripts with categories based on at least metadata, generate information for at least a portion of a plurality of connections based on transcripts, metadata, and categories, and provide the information to a supervisor of agents. The system collects data through video and voice calls. The system can create transcripts of conversations with a module based on artificial intelligence infrastructure. At the same time, it can perform sentiment analysis based on natural language processing methods. Brief Description of the Invention The purpose of this invention is to realize a system developed to determine the success ranking of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also to predict the potential problems and development areas they may encounter. Detailed Description of the Invention "An Improvement System for Educational Institutions" realized to achieve the purpose of this invention is shown in the figure below; Figure 1 is the schematic view of the system subject to the invention. The parts in the figure are numbered one by one, and the equivalents of these numbers are given below: 1. System 2. Application 3. Database 4. Server A system developed to determine the success ranking of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also to predict the potential problems and development areas they may encounter (1); - at least one application (2) conducted on an electronic device, configured to provide an interface containing at least surveys containing questions about students' educational institutions and the education they receive, - recording the survey questions presented to students through the application (2) and the answers given by the students. -exchanging data with the application (2) using any remote communication protocol, recording the students' answers to the survey questions in the database (3) through the communication established with the application (2), via video call Taking the student's answers to the questions, converting them into text and saving them in the database (3), taking the updated data dictionary data from external servers and storing them in the database (3), performing sentiment analysis by taking all the data stored in the database (3) and transferring them to the database (3) on an educational institution basis. It includes at least one server (4) configured to identify and record possible problems and improvements and share them with institutions. The application (2) included in the system (1) that is the subject of the invention is configured to provide an interface that allows students to view and answer survey questions about the education system, social environment and educational institutions by logging in with at least the information defined for them. In one embodiment of the invention, the application (2) is configured to provide an interface to allow students to create personal profiles and write additional comments. The database (3) located in the system (1) that is the subject of the invention is configured to record the student's answers to the questions on the application (2), which the server (4) receives from the application (2), thanks to the communication between the application (2) and the server (4). The database (3) is configured to record the current data dictionary data received by the server (4) from external servers, the results of the analysis made by the server (4) on the answers given to the surveys, and the suggestions determined by the server (4). The server (4) in the system (1) subject to the invention is configured to access the data on the database (3) and to save data on the database (3). The server (4) is configured to regularly record the audio of the conversation with the students, convert it from voice to text with the speech to text (STT) method, and save the texts to the database (3). The server (4) is configured to present surveys through the application (2) in which students' opinions are asked on certain topics at certain times and these opinions are received as text, and to record the answers received in the database (3). The server (4) is configured to transmit questions to the educational institution to measure students' problems and satisfaction and to record the answers received in the database (3). The server (4) is configured to receive the current data dictionary data, including dictionary contents and scores, from external servers and save it to the database (3). The server (4) is configured to apply NLP (Natural Language Processing Method) to the dictionary data saved in the database (3), students' survey answers and interview data. The server (4) is configured to apply the sentiment analysis method to the NLP applied data and save the resulting data to the database (3). The server (4) is configured to perform supervised machine learning on the data received from all institutions and to predict the problems and dissatisfaction that may arise in the institutions. The server (4) is configured to use a numerical coding method for each word prepared by taking into account the language features in which the language analysis will be made. Situations such as where the word is used in the sentence, whether the word is an adjective or a predicate, or whether it is used in an inverted sentence make a difference in numerical coding. The server (4) is configured to compare the analyzes made within the same text or within more than one text of the same person to determine whether the person's emotional state is normal or whether he or she has switched to a certain emotion after a certain moment. Application of the invention to industry In the system subject to the invention (1), surveys will be organized to collect the opinions of students in public or private educational institutions. The same number of male and female students from each age category will be determined at regular intervals, video interviews will be held and their satisfaction with both the social environment and the education they receive will be measured. Apart from the information obtained from these surveys and interviews, the emotional state of the student giving this information during the interview is important. The sincerity and accuracy of the information when giving the information may mean that they say the exact opposite of what they actually feel. A student who says that he is very satisfied with his teachers and the management of the educational institution may actually have the opposite feelings. This can be understood from the place of the predicate in the person's sentence, the inverted sentence, the short answers, or the longer the answer is longer than necessary. The presenter (4) examines all these data with emotion analysis and reveals all the details such as what the students have problems with, what they are happy with, what is done to obtain more efficient results, and these details are constantly recorded in the database (3) and constantly evaluated. A system (1) that educates and develops is being created. In this way, recommendations are made to educational institutions by creating a system (1) that educates with all new developments and the tendencies of students of all ages. The system of the invention (1) determines the success ranking of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also enables the prediction of potential problems and development areas they may encounter. Around these basic concepts, it is possible to develop a wide variety of applications related to the subject of the invention, "An Improvement System for Educational Institutions (1)", and the invention cannot be limited to the examples explained here, it is essentially as stated in the claims. TR TR

Claims

1.STEMS. Developed to determine the success rankings of educational institutions by performing sentiment analysis on surveys and video interviews conducted with students in line with the approvals received, and also to predict the potential problems and development areas they may encounter; - at least one application (2) conducted on an electronic device, configured to provide an interface containing at least surveys containing questions about students' educational institutions and the education they receive, - recording the survey questions presented to students through the application (2) and the answers given by the students. -exchanging data with the application (2) using any remote communication protocol, recording the answers given by the students to the survey questions in the database (3) through the communication established with the application (2), video calling by taking the answers given by the student to the questions, converting them into text and saving them in the database (3), taking the current data dictionary data from external servers and storing them in the database (3), performing sentiment analysis by taking all the data stored in the database (3) and transferring the educational institution to the database (3). A system (1) characterized by at least one server (4) configured to identify and record problems and improvements that may be encountered on a daily basis and share them with institutions. . A system (1) as in claim 1, characterized by an application (2) configured to provide an interface that allows students to view and answer survey questions about the education system, social environment and educational institutions by logging in with at least the information defined for them. ). . A system (1) as in claim 1 or 2, characterized by the application (2) configured to provide an interface to allow students to create personal profiles and write additional comments. . A system as in any of the above claims, characterized by the database (3) configured to record the answers given by the student to the questions on the application (2), which the server (4) receives from the application (2), thanks to the communication between the application (2) and the server (4). As in any of the above claims, characterized by the database (3) configured to record the current data dictionary data received by the server (4) from external servers, the analysis results made by the server (4) on the answers given to the surveys, and the suggestions determined by the server (4). a system (1). . A system (1) as in any of the above claims, characterized by a server (4) configured to access data on the database (3) and save data on the database (3). . A system (1) as in any of the above claims, characterized by a server (4) configured to regularly record the voice of the conversation with the students, convert it from voice to text with the speech to text method, and save the texts to the database (3). . A system as in any of the above claims, characterized by the server (4) configured to present the surveys through the application (2), in which students' opinions are asked on certain subjects at certain times, and these opinions are received as text, and to record the answers received in the database (3), for the purpose of measuring the problems and satisfaction of the students. A system (1) as in any of the above claims, characterized by a server (4) configured to transmit the questions to the educational institution and record the received answers in the database (3). A system (1) as in any of the above claims, characterized by a server (4) configured to receive current data dictionary data, including dictionary contents and scores, from external servers and save it to the database (3). A system as in any of the above claims, characterized by the server (4) configured to apply NLP to the dictionary data saved in the database (3), students' survey answers and interview data, applying the sentiment analysis method to the NLP applied data and recording the obtained data in the database (3). A system as in any of the above claims, characterized by the server (4) configured to perform supervised machine learning on the data received from all institutions and to predict the problems and satisfaction that may arise in the institutions. (one). A system (1) as in any of the above claims, characterized by the server (4) configured to use a numerical coding method for each word prepared by taking into account the language features in which the language analysis will be made. 15. A system as in any of the above claims, characterized by the server (4) configured to determine whether the person's emotional state is normal or whether he or she has switched to a certain emotion after a Definite Moment by comparing the analyzes made within the same text or within more than one text of the same person ( one). TR TR