Karaca Zuccaciye A.S., Istanbul, Turkey: Sentimentality

Written answers to the questions

Total 199 words used.

Karaca is Turkey's leading brand serving in the tableware, kitchen and small electrical appliances categories. It also provides services in the global arena with its more than 300 stores and approximately 4000 employees. At the same time, it gets 1/3 of its turnover from its digital channels. It provides service in Germany, England and France both with its stores and digital platforms. It also sells products on marketplaces in many countries, including America. Our Digital Transformation Unit carries out all the innovations we realize on our website and in our Karaca mobile application, which has managed to break many grounds in its field even though it is only 1.5 years old. Our Digital Transformation Unit consists of 7 sub-units: Data Analytics, Strategy, UX/UI, Design, Digital Marketing, Technology and Growth. As the Digital Transformation Unit, we realize our projects as a teamwork. This project was specially handled by Gamze Kaplan from the Data Analytics Unit.  During the formation of the project, Merve Ekşi, Customer Analytics Manager from the Data Analytics unit, helped with all technical. İrem Karakaş, helped with the application processes for the competition after the project was realized. İrem works for Karaca as Digital Channels Strategy Consultant.

Total 237 words used.

Customers share their experiences after shopping through various channels. This feedback, sometimes on social media, sometimes on product comments, and sometimes on a complaint filed with the call center, includes positive or negative thoughts of the customer about the shopping experience or product features. The process of transforming the text, which is included in these feedbacks and shows the development areas for both the product and the company, into information begins with the collection of data. Comments on the sites of various brands such as Karaca, Emsan, Homend, Karaca-Home, Kaşmir, customer comments on purchases made from the Karaca mobile application, comments made on domestic and international marketplace sites, records created through the call center, return reasons for customers returning their products, and social media. These shares made on the website are collected in the database.

All collected feedback is firstly sentiment analysis done by an artificial intelligence that classifies the positive or negative state of the experience conveyed by the customer. While doing this analysis, it is classified by an artificial intelligence that can classify Turkish or English according to the feedback language. Correction of spelling mistakes made before this classification with normalization is applied. After negative comments, it is classified according to 6 different complaint categories with an artificial intelligence. During the preparation of this classification artificial intelligence, various machine learning and deep learning models were tried and the results of 56 different scenarios were examined.

Total 230 words used.

With this Sentimentality process, using the machine learning  TFIDF Vectorizer algorithm, the words with the highest 2000 document frequency values ​​were selected from the vectorized words. Then, machine learning models were tested with the help of the matrix created using these words.The models were first trained using the default parameters with these matrices. During this training process, 527,362 positive feedbacks and 276,346 negative feedbacks were used.During the evaluation of deep learning models, LSTM, RNN and BERT models were tested. Embedding processes have been done in order to give words as input to deep learning models. Two different approaches were applied during the embedding trials. In the first approach, the Embedding layer of a pre-trained BERT model is used, while in the second approach, the model's own Embedding layer is created.Deep learning outputs include various analyzes such as the trend of product reviews, comments received by products, product sales, and complaint analyzes for suppliers from which these products are purchased, within a decision support system. This report is shared with the entire company. There are also customer complaints within the processed comments for the products. These complaints include both a negative customer experience and point to improvement points for the company. In this way, both product-related problems are solved as soon as they start, and the problem is eliminated at the source by informing the supplier.

Total 221 words used.

• All feedbacks are collected, processed and reported on a single platform.

• Since the whole process is carried out by artificial intelligence, the workload required to classify these feedbacks is reduced.

• It is ensured that the products are defined not only according to the descriptions written on them, but also with their experiences in the eyes of the customer.

• It is ensured that the brand perception in the eyes of the customer is transformed into a measurable KPI.

• Thanks to the alarm system created thanks to the comments processed regularly, problems are observed as soon as they start to occur and are eliminated with agile approaches.

With this project, problems with the products can be solved at the source. One of the biggest outputs of this quality increase is the decrease in return rates. Thanks to the decrease in the return rates, an additional income of 7 Million TL per month , 84M TL per year can be provided.

Thanks to the fact that artificial intelligence manages the whole process, an average of 30 thousand weekly feedback data is processed automatically from end to end.

Thanks to this automation, the weekly workload of 5 people required to read comments is improved. Due to the annual productivity of this project 1.8 Million TL from man/hour expenditures are gained.