-
Time Series Clustering for Grouping Products Based on Price and Sales Patterns
Authors:
Aysun Bozanta,
Sean Berry,
Mucahit Cevik,
Beste Bulut,
Deniz Yigit,
Fahrettin F. Gonen,
Ayşe Başar
Abstract:
Developing technology and changing lifestyles have made online grocery delivery applications an indispensable part of urban life. Since the beginning of the COVID-19 pandemic, the demand for such applications has dramatically increased, creating new competitors that disrupt the market. An increasing level of competition might prompt companies to frequently restructure their marketing and product p…
▽ More
Developing technology and changing lifestyles have made online grocery delivery applications an indispensable part of urban life. Since the beginning of the COVID-19 pandemic, the demand for such applications has dramatically increased, creating new competitors that disrupt the market. An increasing level of competition might prompt companies to frequently restructure their marketing and product pricing strategies. Therefore, identifying the change patterns in product prices and sales volumes would provide a competitive advantage for the companies in the marketplace. In this paper, we investigate alternative clustering methodologies to group the products based on the price patterns and sales volumes. We propose a novel distance metric that takes into account how product prices and sales move together rather than calculating the distance using numerical values. We compare our approach with traditional clustering algorithms, which typically rely on generic distance metrics such as Euclidean distance, and image clustering approaches that aim to group data by capturing its visual patterns. We evaluate the performances of different clustering algorithms using our custom evaluation metric as well as Calinski Harabasz and Davies Bouldin indices, which are commonly used internal validity metrics. We conduct our numerical study using a propriety price dataset from an online food and grocery delivery company, and the publicly available Favorita sales dataset. We find that our proposed clustering approach and image clustering both perform well for finding the products with similar price and sales patterns within large datasets.
△ Less
Submitted 18 April, 2022;
originally announced April 2022.
-
Text Classification for Predicting Multi-level Product Categories
Authors:
Hadi Jahanshahi,
Ozan Ozyegen,
Mucahit Cevik,
Beste Bulut,
Deniz Yigit,
Fahrettin F. Gonen,
Ayşe Başar
Abstract:
In an online shopping platform, a detailed classification of the products facilitates user navigation. It also helps online retailers keep track of the price fluctuations in a certain industry or special discounts on a specific product category. Moreover, an automated classification system may help to pinpoint incorrect or subjective categories suggested by an operator. In this study, we focus on…
▽ More
In an online shopping platform, a detailed classification of the products facilitates user navigation. It also helps online retailers keep track of the price fluctuations in a certain industry or special discounts on a specific product category. Moreover, an automated classification system may help to pinpoint incorrect or subjective categories suggested by an operator. In this study, we focus on product title classification of the grocery products. We perform a comprehensive comparison of six different text classification models to establish a strong baseline for this task, which involves testing both traditional and recent machine learning methods. In our experiments, we investigate the generalizability of the trained models to the products of other online retailers, the dynamic masking of infeasible subcategories for pretrained language models, and the benefits of incorporating product titles in multiple languages. Our numerical results indicate that dynamic masking of subcategories is effective in improving prediction accuracy. In addition, we observe that using bilingual product titles is generally beneficial, and neural network-based models perform significantly better than SVM and XGBoost models. Lastly, we investigate the reasons for the misclassified products and propose future research directions to further enhance the prediction models.
△ Less
Submitted 2 September, 2021;
originally announced September 2021.
-
Experimental Analysis and Evaluation of RaptorQ Codes for Video Multicasting over Wi-Fi
Authors:
Berna Bulut
Abstract:
This paper presents a reliable and efficient high quality video streaming solution for use in challenging outdoor environments over Wi-Fi. An application layer forward error correction based on RaptorQ codes was implemented in a practical Wi-Fi based server and client system to enhance reliability. Thus, this paper presents the first detailed analysis on the implementation of RaptorQ codes for str…
▽ More
This paper presents a reliable and efficient high quality video streaming solution for use in challenging outdoor environments over Wi-Fi. An application layer forward error correction based on RaptorQ codes was implemented in a practical Wi-Fi based server and client system to enhance reliability. Thus, this paper presents the first detailed analysis on the implementation of RaptorQ codes for streaming high definition video over Wi-Fi. The measurements were performed in central Bristol with parameters such as RaptorQ symbol size, code rate, buffering time and modulation and coding scheme, and user quality of experience based on these parameters was evaluated. For multicast live video streaming it is demonstrated that system performance is mostly dominated by hardware and software limitations on constrained host platforms where the incoming packet rate exceeds the device`s ability to consume the traffic, i.e., Wi-Fi clients are a major source of packet loss, even in ideal channel conditions. Client limitations were found to be a function of modulation and coding schemes and RaptorQ coding parameters. Therefore, the optimum system design parameters such as RaptorQ symbol size, code rate and buffering time with respect to modulation and coding schemes were suggested considering practical limitations from real-world measurements.
△ Less
Submitted 26 April, 2020;
originally announced April 2020.
-
Massive Multiple Input Massive Multiple Output for 5G Wireless Backhauling
Authors:
Dinh-Thuy Phan-Huy,
Philippe Ratajczak,
Raffaele D'Errico,
Jan Jarvelainen,
Di Kong,
Katsuyuki Haneda,
Berna Bulut,
Aki Karttunen,
Mark Beach,
Evangelos Mellios,
Mario Castaneda,
Mythri Hunukumbure,
Tommy Svensson
Abstract:
In this paper, we propose a new technique for the future fifth generation cellular network wireless backhauling. We show that hundreds of bits per second per Hertz (bits per second per Hz) of spectral efficiency can be attained at a high carrier frequency (such as 26 GHz) between large antenna arrays deployed along structures (such as lamp posts) that are close and roughly parallel to each other.…
▽ More
In this paper, we propose a new technique for the future fifth generation cellular network wireless backhauling. We show that hundreds of bits per second per Hertz (bits per second per Hz) of spectral efficiency can be attained at a high carrier frequency (such as 26 GHz) between large antenna arrays deployed along structures (such as lamp posts) that are close and roughly parallel to each other. Hundreds of data streams are spatially multiplexed through a short range and line of sight massive multiple input massive multiple output propagation channel thanks to a new low complexity spatial multiplexing scheme, called block discrete Fourier transform based spatial multiplexing with maximum ratio transmission. Its performance in real and existing environments is assessed using accurate ray-tracing tools and antenna models. In the best simulated scenario, 1.6 kbits per second per Hz of spectral efficiency is attained, corresponding to 80% of Singular Value Decomposition performance, with a transmitter and a receiver that are 200 and 10,000 times less complex, respectively.
△ Less
Submitted 12 March, 2018;
originally announced March 2018.
-
Millimeter Wave Channel Measurements in a Railway Depot
Authors:
Berna Bulut,
Thomas Barratt,
Di Kong,
Jue Cao,
Alberto Loaiza Freire,
Simon Armour,
Mark Beach
Abstract:
Millimeter wave (mmWave) communication is a key enabling technology with the potential to deliver high capacity, high peak data rate communications for future railway services. Knowledge of the radio characteristics is of paramount importance for the successful deployment of such systems. In this paper mmWave channel measurements are reported for a railway environment using a wideband channel soun…
▽ More
Millimeter wave (mmWave) communication is a key enabling technology with the potential to deliver high capacity, high peak data rate communications for future railway services. Knowledge of the radio characteristics is of paramount importance for the successful deployment of such systems. In this paper mmWave channel measurements are reported for a railway environment using a wideband channel sounder operating at 60GHz. Highly directional antennas are deployed at both ends of the link. Data is reported for path loss, root mean square (RMS) delay spread and K-factor. Static and mobile measurements are considered. Analysis shows that the signal strength is strongly dependent (up to 25dB) on the azimuth orientation of the directional transmit and receive antennas. A path loss exponent of n=2.04 was extracted from the Line-of-Sight measurements with optimally aligned antennas. RMS delay spreads ranged from 1ns to 22ns depending on antenna alignment. 50% of the measured K-factors were found to be less than 6dB. We conclude this is the result of ground reflections in the vertical Tx-Rx plane.
△ Less
Submitted 30 June, 2018; v1 submitted 16 August, 2017;
originally announced August 2017.
