Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Environmental Sciences (1081)
- Computer Sciences (463)
- Chemistry (449)
- Engineering (156)
- Computer Engineering (150)
-
- Polymer Chemistry (141)
- Mathematics (120)
- Statistics and Probability (58)
- Earth Sciences (56)
- Physics (56)
- Geology (32)
- Life Sciences (16)
- Biotechnology (15)
- Artificial Intelligence and Robotics (11)
- Education (9)
- Educational Assessment, Evaluation, and Research (9)
- Databases and Information Systems (6)
- Chemical Engineering (4)
- Business (2)
- Catalysis and Reaction Engineering (1)
- Environmental Health (1)
- Environmental Public Health (1)
- Intellectual Property Law (1)
- Law (1)
- Management Information Systems (1)
- Medicine and Health Sciences (1)
- Operations Research, Systems Engineering and Industrial Engineering (1)
- Pharmacology, Toxicology and Environmental Health (1)
- Programming Languages and Compilers (1)
- Keyword
- Publication Year
- Publication
- Publication Type
Articles 301 - 330 of 2268
Full-Text Articles in Physical Sciences and Mathematics
Adaptive Image Preprocessing And Augmentation For Disease Screening On Multi-Source Chest X-Ray Datasets, Wasunan Chokchaithanakul
Adaptive Image Preprocessing And Augmentation For Disease Screening On Multi-Source Chest X-Ray Datasets, Wasunan Chokchaithanakul
Chulalongkorn University Theses and Dissertations (Chula ETD)
Research on deep learning models for chest radiology applications has increased attention by the public. However, most works focus on developing models using in-domain data, so the significant drawback, when applied in real-world scenarios, was the mismatched data with the training set. Consequently, some models perform inferior at the deployment stage. This work focused on the effects of dataset mismatch on chest radiography and analyzed the methods the overcome the mismatch issues. The lung balance contrast enhancement technique (lung BCET) automatically identifies the lung region and normalizes the image accordingly to improve the robustness of out-of-domain data developed. Additionally, augmentation …
Genetic Code Expansion Of Plastic-Degrading Enzymes For Detection Of Microplastics, Jariya Jitdee
Genetic Code Expansion Of Plastic-Degrading Enzymes For Detection Of Microplastics, Jariya Jitdee
Chulalongkorn University Theses and Dissertations (Chula ETD)
Petrochemical plastics can degrade into the size level of microplastics. One source of microplastics is polyethylene terephthalate (PET or PETE). Enzymes used to degrade PET have been reported. In our work, we wish to repurpose the activity of PET-degrading enzymes, into PET-detecting enzymes, and use the engineered enzyme for microplastic detection, via enzyme engineering to create enzyme variants which can form covalent adducts with microplastic particles. The enzyme-microplastic adduct is generated via the use of 2,3-diaminopropionic acid (DAP) derivative, which we will incorporate in place of the catalytic serine residue at the active site of I. sakeinesis—PETase via the genetic …
Chemical Constituents Of Fan-Si Abutilon Indicum Stems, Arum Restu Widyasti
Chemical Constituents Of Fan-Si Abutilon Indicum Stems, Arum Restu Widyasti
Chulalongkorn University Theses and Dissertations (Chula ETD)
This study aimed to isolate secondary metabolites from the stems of Abutilon indicum and to evaluate their α-glucosidase, pancreatic lipase, α-chymotrypsin inhibitory activity. Chromatographic fractionation of the n-hexane crude extracts led to the isolation of three known metabolite (1-3). The crude extract from all parts of the plant were evaluated for inhibitory activity on the alpha-glucosidase enzyme. The DCM-soluble roots were active with 10.2% at 25 μg/mL. The hexane and MeOH-soluble stems were active with 11.0% at 25 μg/mL and 8.0% at 0.25 μg/mL. The hexane, DCM, and MeOH-soluble leaves are active with 20.2% at 0.025 μg/mL; 20.0% at 0.025 …
Hydrogen Gas Sensing And Storage Properties Of Doped And Non-Doped Beryllium Oxide Nanotubes, Kritsanaphas Sawing
Hydrogen Gas Sensing And Storage Properties Of Doped And Non-Doped Beryllium Oxide Nanotubes, Kritsanaphas Sawing
Chulalongkorn University Theses and Dissertations (Chula ETD)
The electronic properties of armchair (5,5) beryllium oxide nanotube (BeONT), zigzag (10,0) BeONT, and their surfaces doped by the selected elements of periods 2 (B, C and N), 3 (Mg, Al, Si, P and S) and 4 (Ca, Ga, Ge, As and Se), and their hydrogen molecule adsorptions were studied using periodic DFT‒D method. The C(O)‒ and Ge(O)‒(5,5) BeONTs, C(O)‒, Si(O)‒ and Ge(O)‒(10,0) BeONTs were suggested to be new hydrogen molecule sensing materials based on the electrical resistivity measurement. The C(O)‒, Si(O)‒ and Ge(O)‒doped (5,5) BeONTs, and C(O)‒ and Si(O)‒doped (10,0) BeONTs were suggested to be utilized as hydrogen storage …
Colorimetric And Fluorescent Detection Of Nucleic Acid By Styryl Dyes, Kotchakorn Supabowornsathit
Colorimetric And Fluorescent Detection Of Nucleic Acid By Styryl Dyes, Kotchakorn Supabowornsathit
Chulalongkorn University Theses and Dissertations (Chula ETD)
Nucleic acid staining dyes are essential tools for the analysis and visualizing of DNA/RNA in vitro and in cellular applications. Although there are several commercially available dyes developed during the past few decades, the selection is still relatively limited, and they are often very costly and associated with undesirable characteristics such as toxicity. Consequently, the discovery of nontoxic, readily available dyes, with desirable and controllable optical characteristics remains important. Styryl dyes have recently gained popularity as potential biological staining agents with many desirable properties including a straightforward synthesis procedure, excellent photostability, tunable optical properties, and high responsiveness towards nucleic acid …
Synthesis Of Quantum Dot-Based Fluorescent Sensors For Detection Of Anionic Metabolites And Copper (Ii) Ion, Nattakarn Phromsiri
Synthesis Of Quantum Dot-Based Fluorescent Sensors For Detection Of Anionic Metabolites And Copper (Ii) Ion, Nattakarn Phromsiri
Chulalongkorn University Theses and Dissertations (Chula ETD)
Oxalate, citrate, and urate are the anionic metabolites that are the products of the metabolism pathway of living organisms. The high level of these anionic metabolites is related to the diseases. For example, the excess amount of oxalate ions in urine indicates the risk of kidney stones, while the level of citrate can suggest the risk of cancer. The level of urate in urine is related to the risk of gout. The detection of these anionic metabolites is significant and still remains a challenge. Thus, this work aims to develop the more sensitive and selective fluorescence sensing of oxalate in …
Boron Nanomaterials And Small Molecules Adsorption, Nontawat Ploysongsri
Boron Nanomaterials And Small Molecules Adsorption, Nontawat Ploysongsri
Chulalongkorn University Theses and Dissertations (Chula ETD)
The adsorption abilities of hydrogen boride nanotubes and nanosheet were studied. The adsorption of H2, H2O, NH3 and CH4 on the most stable armchair and zigzag hydrogen boride nanotubes and their C, N and O decorated nanotubes were investigated by using density functional tight binding (DFTB) method. The results show that C-decorated armchair and zigzag HBNTs could be the NH3 storage materials and N-decorated armchair and zigzag HBNTs could be water vapor sensing materials. The adsorption of Li, Na and K atoms on armchair like (5,5) and zigzag like (10,0) HBNTs and hydrogen adsorption on Li, Na and K decorated …
Development Of Retention Index Based Simulation For Validation Of Compound Identification In Gc×Gc, Palathip Kakanopas
Development Of Retention Index Based Simulation For Validation Of Compound Identification In Gc×Gc, Palathip Kakanopas
Chulalongkorn University Theses and Dissertations (Chula ETD)
Comprehensive two-dimensional gas chromatography (GC×GC) is a high-performance technique for separation, identification and quantification of volatiles and semi-volatiles in complex multi-component samples such as biomolecular molecules, essential oil, foods, and petroleum. One of the most popular detectors used for peak identification with GC×GC is mass spectrometer (MS) allowing identification of separated peaks based on comparison with mass spectral library. However, only MS library comparison shows low confidence in compound identifications due to the fact that compounds with similar structures (especially for isomers) often have similar mass spectra. Apart from sample preparation, a great challenge is to effectively select types of …
Thin-Layer Chromatography For Determination Of Phenolic Compound Profile And Antioxidant Activity Of Thai Honey, Pattraporn Chobpradit
Thin-Layer Chromatography For Determination Of Phenolic Compound Profile And Antioxidant Activity Of Thai Honey, Pattraporn Chobpradit
Chulalongkorn University Theses and Dissertations (Chula ETD)
In this study, a small-scale method using thin-layer chromatography (TLC) was developed to determine profile and antioxidant activity value of Thai honey samples obtained from different flora sources. Prior to TLC analysis, the samples were prepared by solvent extraction with dichloromethane. Using 2, 2-diphenyl-1-picrylhydrazy radical (DPPH•) reacted with the samples extract prior to TLC spot and color detection with imageJ, half-maximal inhibitory concentration (IC50) values of seven honey samples was obtained in a range of 9 to 22 mg/mL in comparison with IC50 values of 0.13 mg/mL for L-ascorbic acid (LA), along with limit of detection (LOD) and limit of …
Theoretical Study Of Ethylene Polymerization By Phenoxy-Imine Catalyst, Pavee Apilardmongkol
Theoretical Study Of Ethylene Polymerization By Phenoxy-Imine Catalyst, Pavee Apilardmongkol
Chulalongkorn University Theses and Dissertations (Chula ETD)
Reaction mechanisms of ethylene polymerization catalyzed by the phenoxy-imine (FI) and the nickel phenoxyphosphine polyethylene glycol (Ni-PEG) with alkali metals were explored using DFT calculations. For FI catalysts, the effect of group IVB transition metals substitutions was investigated. The trend of calculated activation energies (Ea) at the rate-determining step is Zr < Hf < Ti and is in good agreement with experiments. The effect of ligands of the Ti-FI-based catalysts when changing the parent nitrogen (O, N) to oxygen (O, O), phosphorus (O, P), and sulfur (O, S) ligands on activity was also monitored. The results indicated that the sulfur (O, S) ligand gives the lowest activation energy. Additionally, the reactivity of Ni-phenoxy-imine (Ni-FI)-based catalysts for polyethylene polymerization was studied. Our calculations suggested that the square planar complex of Ni-FI is more reactive than its C2 symmetric octahedral complex. For Ni-PEG(M) catalysts, the trend for activation energies of four Ni-PEG(M) systems is Li < Na < K < Cs, which corresponds to experimentally observed activities. Moreover, the roles of secondary metals in Ni-PEG catalysts in terms of steric, electronic, and electrostatic effects were elucidated. The DFT results suggested that the active catalyst should have strong cooperative metal-metal/metal-ligand interactions and less positive charge on the secondary metal. Finally, to gain insight into the design of the novel Ni-PEG catalysts with alkali-earth metals, the effect of catalyst structure on experimental activity was investigated. This work provides fundamental understandings of the reaction mechanisms for the FI and Ni-PEG(M) catalysts, which could be used for the design and development of catalysts for ethylene polymerization.
Discrimination Of Weedy Rice By Using Near-Infrared Spectroscopy Combined With Chemometrics, Sureerat Makmuang
Discrimination Of Weedy Rice By Using Near-Infrared Spectroscopy Combined With Chemometrics, Sureerat Makmuang
Chulalongkorn University Theses and Dissertations (Chula ETD)
Weedy rice is one of the most notorious weeds occurring in rice-growing areas, especially in South-East Asia. Weedy rice especially in form of paddy seed is difficult to manage and separate as they provide common features (morphological resemblance) to cultivated rice. This work presents a modification of self-organizing map (SOMs) for the classification of weedy rice from cultivated rice via in situ direct sample analysis from paddy seed using near-infrared (NIR) spectroscopy and hyperspectral NIR camera. The sample pretreatment was carried out by a cyclone vacuum machine to remove the contaminated particles and other impurities. The physical characteristics and the …
Program Development For Synthesis Plan Design Of Organic Compounds By Using Artificial Intelligence, Tawatchai Jitporn
Program Development For Synthesis Plan Design Of Organic Compounds By Using Artificial Intelligence, Tawatchai Jitporn
Chulalongkorn University Theses and Dissertations (Chula ETD)
Computer-Assisted Synthesis Planning (CASP) is a computational tool for facilitating organic chemists to design and synthesis organic compounds by calculating the best possible synthesis pathways. Most CASP tools are commercial or closed-source software and thus, they cannot be modified to improve performance or update database. Therefore, it is our aim to develop an open-source CASP tool. Library of reaction rules that was created from the US patent database was taken from literature and was then reclassified to improve its applicability. This modified library was then used for training neural network and 2 important models were obtained for the search engine. …
Flow-Through Reactor For Photomediated Synthesis Of Silver Nanoprisms, Thanasit Laisanguanngam
Flow-Through Reactor For Photomediated Synthesis Of Silver Nanoprisms, Thanasit Laisanguanngam
Chulalongkorn University Theses and Dissertations (Chula ETD)
Silver nanoprisms (AgNPrs) have been widely used due to their outstanding optical properties which is one of an important phenomenon that depends on their size, shape, and compositions with characteristic localized surface plasmon resonance (LSPR). The changes in LSPR of AgNPrs can be simply observed by the color changes of the AgNPrs colloidal solution, therefore, the AgNPrs have been mostly used as chemical sensors as they provide high sensitivity and capable to be functionalized to improve the detection selectivity. In this study, we report a photochemical technique to induce the shape conversion of the spherical silver nanoparticle to AgNPrs by …
Development Of Colorimetric Paper-Based Analytical Device For Cortisol Detection, Thanathip Kosawatphat
Development Of Colorimetric Paper-Based Analytical Device For Cortisol Detection, Thanathip Kosawatphat
Chulalongkorn University Theses and Dissertations (Chula ETD)
Nowadays, the detection of cortisol can play a key role in understanding stress-related diseases. Several competitive immunoassay formats have been reported for the detection of this small hormone molecule. Lateral flow immunoassay (LFIA) is one of the most widely used platforms that offer its simplicity and rapidity for such analysis. Unlike previous sensors for cortisol sensing that require complicated readers, specialized reagents, or even post-treatment methods to amplify the obtained signal, in this work, we employed a concave test zone constructed on the LFIA device (cLFIA) which can directly enhance the signal response as the flowing sample was substantially concentrated …
Incorporating Context Into Non-Autoregressive Model Using Contextualized Ctc For Sequence Labelling, Burin Naowarat
Incorporating Context Into Non-Autoregressive Model Using Contextualized Ctc For Sequence Labelling, Burin Naowarat
Chulalongkorn University Theses and Dissertations (Chula ETD)
Connectionist Temporal Classification (CTC) loss has become widely used in sequence modeling tasks such as Automatic Speech Recognition (ASR) and Handwritten Text Recognition (HTR) due to its ease of use. CTC itself has no architecture constraints, but it is commonly used with recurrent models that predict letters based on histories in order to relax the conditional independent assumption. However, recent sequence models that incorporate CTC loss have been focusing on speed by removing recurrent structures, hence losing important context information. This thesis presents Contextualized Connectionist Temporal Classification (CCTC) loss, which induces prediction dependencies in non-recurrent and non-autoregressive neural networks for …
Voice Impersonation For Thai Speech Using Cyclegan Over Prosody, Chatri Chuanngulueam
Voice Impersonation For Thai Speech Using Cyclegan Over Prosody, Chatri Chuanngulueam
Chulalongkorn University Theses and Dissertations (Chula ETD)
No abstract provided.
Data Augmentation For Thai Natural Language Processing Using Different Tokenization, Patawee Prakrankamanant
Data Augmentation For Thai Natural Language Processing Using Different Tokenization, Patawee Prakrankamanant
Chulalongkorn University Theses and Dissertations (Chula ETD)
Tokenization is one of the most important data pre-processing steps in the text classification task and also one of the main contributing factors in the model performance. However, getting good tokenizations is non-trivial when the input is noisy, and is especially problematic for languages without an explicit word delimiter such as Thai. Therefore, we proposed an alternative data augmentation method to improve the robustness of poor tokenization by using multiple tokenizations. We evaluated the performance of our algorithms on different Thai text classification datasets. The results suggested our augmentation scheme makes the model more robust to tokenization errors and can …
Spectral And Latent Representation Distortion For Tts Evaluation, Thananchai Kongthaworn
Spectral And Latent Representation Distortion For Tts Evaluation, Thananchai Kongthaworn
Chulalongkorn University Theses and Dissertations (Chula ETD)
One of the main problems in the development of text-to-speech (TTS) systems is its reliance on subjective measures, typically the Mean Opinion Score (MOS). MOS requires a large number of people to reliably rate each utterance, making the development process slow and expensive. Recent research on speech quality assessment tends to focus on training models to estimate MOS, which requires a large number of training data, something that might not be available in low-resource languages. We propose an objective assessment metric based on the DTW distance using the spectrogram and the high-level features from an Automatic Speech Recognition (ASR) model …
Demand Forecasting In Production Planning For Dairy Products Using Machine Learning And Statistical Methods, Chayuth Vithisoontorn
Demand Forecasting In Production Planning For Dairy Products Using Machine Learning And Statistical Methods, Chayuth Vithisoontorn
Chulalongkorn University Theses and Dissertations (Chula ETD)
Demand forecasting is an essential task in manufacturing of every industry. Efficient forecasting relieves the excessive stock and out-of-stock problem, reducing revenue loss. This research performs a direct multistep forecast approach of demand forecasting on 8 dairy products of 5 different dairy production plants with 5-year data. Widely used traditional statistical method and the state of the art deep learning method for sequence problems are picked. ARIMA and LSTM. The models are compared in many aspects, monthly observations against weekly observations, univariate against multivariate, and statistical against deep learning using model error and business metrics. The result shows that both …
Conformance Checking And Discovery Of Information Service Request Process, Liam Khaosanoi
Conformance Checking And Discovery Of Information Service Request Process, Liam Khaosanoi
Chulalongkorn University Theses and Dissertations (Chula ETD)
Process mining is a form of business process analysis. The approach could support organizations in retrieving structured process information by using recorded process data to discover, monitor and improve the processes. In this work, the process mining technology is applied for conformance checking and discovering an organization’s Information Service Request process in reality as a case study. Currently, the reference model was drawn as a simple flow using Microsoft Word. The proposed method starts with writing a VBA script to extract traces from the drawing that enables the generation of XES file used for modeling the reference process in Petri …
Object Detection In Intelligent Billing System For Conveyor Belt Sushi Restaurant, Rangrak Maitriboriruks
Object Detection In Intelligent Billing System For Conveyor Belt Sushi Restaurant, Rangrak Maitriboriruks
Chulalongkorn University Theses and Dissertations (Chula ETD)
Organization must automate wherever and whenever they can, particularly during today’s global changes in daily lifestyles. Trends regarding the use of technology, especially AI has emerged as a key enabler for disruptive innovation. This thesis thus presents the application programming interface of object detector implemented with YOLOv4 and OpenCV for classifying the prices of sushi plates distinguished by colors. The object detector is part of the smart cross-platform mobile application to facilitate billing process for conveyor belt sushi business. The frontend is developed with Flutter to build single codebase for UIs. To handle the variants of image colors resulting from …
A Deep Learning Model For Predicting Long Non-Coding Rna And Messenger Rna With Model Interpretation, Rattaphon Lin
A Deep Learning Model For Predicting Long Non-Coding Rna And Messenger Rna With Model Interpretation, Rattaphon Lin
Chulalongkorn University Theses and Dissertations (Chula ETD)
Long non-coding RNAs (lncRNAs) play important roles in many biological processes and are found to be associated with several diseases. The development of next-generation sequencing technologies has discovered numerous unannotated transcripts. However, classifying these unannotated transcripts by using biological experiments is very time-consuming and expensive. Thus, a computational approach is considered as an alternative solution which is faster and cheaper. Many existing lncRNA identification tools are available, these tools lack an explanation of which features contributed to their prediction results. Here, we present Xlnc1DCNN, a tool for distinguishing long non-coding RNAs (lncRNAs) from protein-coding transcripts (PCTs) together with a prediction …
A Design Of Fpga Framework For Quantum Computing Simulation, Yaninee Jungjarassub
A Design Of Fpga Framework For Quantum Computing Simulation, Yaninee Jungjarassub
Chulalongkorn University Theses and Dissertations (Chula ETD)
We use FPGA to optimize the simulation of quantum computing in two aspects. (a) The if-else state is used in place of tensor product calculation. This allows the tensor product of each quantum operator to be generated in a single clock cycle. (b) The pre-calculated lookup ROM is used for estimating the sine and cosine values. This facilitates the computation of quantum gates that are related to angle. To validate our work, we implement our design in VerilogHDL. The performance is evaluated using an FPGA simulator. The result shows a dramatic improvement in the simulation process comparing to those of …
Cascading Model For Forex Market Forecasting Using Fundamental And Technical Indicator Data Based On Bert, Arisara Pornwattanavichai
Cascading Model For Forex Market Forecasting Using Fundamental And Technical Indicator Data Based On Bert, Arisara Pornwattanavichai
Chulalongkorn University Theses and Dissertations (Chula ETD)
The foreign exchange rate market is the world's biggest and most liquid financial market, and it's where all currency pairs' exchange rates are set. Since foreign exchange (Forex) rates play a critical role in financial technology and business, many researchers are now interested in forecasting them. The characteristics of Forex data, that include fluctuation, non-linearity, and random walk phenomena, make it difficult for forecasting. Several related studies integrate fundamental data (FD) and technical indicator data to generate Forex forecasting signals (TI). TI is a price pattern-based signal, whereas FD is an indicator of the country's economic conditions. Nevertheless, when it …
Pulmonary Lesion Classification Using Convolutional Neural Network For Endobronchial Ultrasonogram, Banphatree Khomkham
Pulmonary Lesion Classification Using Convolutional Neural Network For Endobronchial Ultrasonogram, Banphatree Khomkham
Chulalongkorn University Theses and Dissertations (Chula ETD)
This dissertation aims to develop a method to help classify pulmonary lesions from endobronchial ultrasonography images by proposing new features that are extracted from an EBUS image based on medical knowledge and a pulmonary lesion classification framework. The proposed features, namely the adaptive weighted-sum of the upper triangular gray-level co-occurrence matrix and the adaptive weighted-sum of the lower triangular gray-level co-occurrence matrix are used to determine heterogeneity, which is one of the most important characteristics of malignancy. The proposed features together with other standard features are used as input data for the proposed classification framework that uses the weighted ensemble …
Travel Time Prediction With Graph Neural Network: A Case Study In Bangkok Thailand, Sathita Buapang
Travel Time Prediction With Graph Neural Network: A Case Study In Bangkok Thailand, Sathita Buapang
Chulalongkorn University Theses and Dissertations (Chula ETD)
Traffic prediction is an essential and challenging task for traffic management and commercial purposes. Machine learning methods for traffic prediction usually treat traffic conditions as time-series due to obvious temporal patterns. Recently, spatial relationships among roads in a road network have also been used to improve traffic prediction. This study proposes a novel method to predict traffic conditions such as speed using a graph convolutional neural network with a spectral adjacency matrix (GCN-Spectral). Unlike a spatial adjacency matrix representing physical connections between road segments, a spectral matrix represents the correlation between road segments regarding traffic conditions. The GCN-Spectral model is …
Classification Of Abusive Thai Messages In Social Networks Using Deep Learning, Ruangsung Wanasukapunt
Classification Of Abusive Thai Messages In Social Networks Using Deep Learning, Ruangsung Wanasukapunt
Chulalongkorn University Theses and Dissertations (Chula ETD)
Social media has improved on traditional news sources by allowing increased access to information. However, the anonymity social media provides can lead to abusive and hateful speech without detection or repercussion from individuals with malicious intentions. This research develops a binomial and a multinomial classification model for classifying Thai social media text for five categories of abusive content detection in social media that include Rude, Figurative, Dirty, Offensive and Non-Abusive. The experiments demonstrated that DistilBERT achieved the highest F1 score with 0.8510 for the binomial model and 0.9067 for the multinomial model. BiLSTM performed second best with an F1 score …
Considering Neighbor Projection On Neural Based Recommender Systems, Thitiporn Neammanee
Considering Neighbor Projection On Neural Based Recommender Systems, Thitiporn Neammanee
Chulalongkorn University Theses and Dissertations (Chula ETD)
Now, CF is applied with a neural network to make the model more flexible and more accurate. Different neighbors have a different influence on the target user, and different users usually have different rating patterns. Therefore, the proposed method needs to consider two major issues when applying CF with a neural network: the similarity levels between the neighbors and the target user and the user's rating pattern conversion. Thus, the proposed method consists of three main modules to solve the issues mentioned above: rating conversion, similarity module uses, and prediction module. In the experiment, the proposed method is evaluated and …
Thai Variable-Length Question Classification For E-Commerce Platform Using Machine Learning With Topic Modeling Feature, Wasu Chunhasomboon
Thai Variable-Length Question Classification For E-Commerce Platform Using Machine Learning With Topic Modeling Feature, Wasu Chunhasomboon
Chulalongkorn University Theses and Dissertations (Chula ETD)
Nowadays, e-commerce platform continuously grows every year and becomes a part of our daily life. However, the application changes from time to time. Either new users or experienced users could face a problem. Several channels, which are FAQ, email, live chat, and call, are provided by e-commerce platform to cope with the problem. FAQ is usually ignored because it is hard to search for the desired answer. The rest channels are applicable. However, the huge number of users causes a bottleneck especially in the special events which delays the users to receive help because customer service agent can reply to …
Natural Language Processing For Digital Advertising, Yiping Jin
Natural Language Processing For Digital Advertising, Yiping Jin
Chulalongkorn University Theses and Dissertations (Chula ETD)
Advertising is not only a marketing or sales activity but a particular form of two-way communication. In this thesis, we propose to apply the two main subtasks of natural language processing (NLP), namely natural language understanding (NLU) and natural language generation (NLG), to digital advertising to enhance the effectiveness of advertising. We apply weakly-supervised text classification to rapidly build text classifiers for contextual advertising (Jin et al. 2022). The method requires a handful of labeled keywords instead of a large corpus of labeled documents and can be easily transferred to new domains. We further evaluate the weakly-supervised models using unsupervised …