Search results for: stevia rebaudiana bert

Authors: T. Bettaieb, S. Soufi, S. Arbaoui

Abstract:

Stevia rebaudiana Bert. is a natural sweet plant. The leaves contain diterpene glycosides stevioside, rebaudiosides A-F, steviolbioside and dulcoside, which are responsible for its sweet taste and have commercial value all over the world as sugar substitute in foods and medicines. Stevia rebaudiana Bert. is sensitive temperature lower than 9°C. The possibility of its outdoor culture in Tunisian conditions demand genotypes tolerant to low temperatures. In order to evaluate the low temperature tolerance of eight genotypes of Stevia rebaudiana, the activities of superoxide dismutase (SOD), ascorbate peroxidase (APX) and catalases (CAT) were measured. Before carrying out the analyses, three genotypes of Stevia were exposed for 1 month at a temperature regime of 18°C during the day and 7°C at night similar to winter conditions in Tunisia. In response to the stress generated by low temperature, antioxidant enzymes activity revealed on native gel and quantified by spectrophotometry showed variable levels according to their degree of tolerance to low temperatures.

Keywords: chilling tolerance, enzymatic activity, stevia rebaudiana bert, low thermal stress

Procedia PDF Downloads 413

Authors: Vincent Alexander, Rizkita Esyanti

Abstract:

Leaves of Stevia rebaudiana contain steviol glycoside which mainly comprise of stevioside, a natural sweetener compound that is 100-300 times sweeter than sucrose. Current cultivation method of Stevia rebaudiana in Indonesia has yet to reach its optimum efficiency and productivity to produce stevioside as a safe sugar substitute sweetener for people with diabetes. An alternative method that is not limited by environmental factor is in vitro temporary immersion system (TIS) culture method using recipient for automated immersion (RITA^®) bioreactor. The aim of this research was to evaluate the effect of red LED light induction towards shoot growth and stevioside accumulation in TIS RITA^® bioreactor system, as an endeavour to increase the secondary metabolite synthesis. The result showed that the stevioside accumulation in TIS RITA^® bioreactor system induced with red LED light for one hour during night was higher than that in TIS RITA^® bioreactor system without red LED light induction, i.e. 71.04 ± 5.36 μg/g and 42.92 ± 5.40 μg/g respectively. Biomass growth rate reached as high as 0.072 ± 0.015/day for red LED light induced TIS RITA^® bioreactor system, whereas TIS RITA^® bioreactor system without induction was only 0.046 ± 0.003/day. Productivity of Stevia rebaudiana shoots induced with red LED light was 0.065 g/L medium/day, whilst shoots without any induction was 0.041 g/L medium/day. Sucrose, salt, and inorganic consumption in both bioreactor media increased as biomass increased. It can be concluded that Stevia rebaudiana shoot in TIS RITA^® bioreactor induced with red LED light produces biomass and accumulates higher stevioside concentration, in comparison to bioreactor without any light induction.

Keywords: LED, Stevia rebaudiana, Stevioside, TIS RITA

Procedia PDF Downloads 339

Authors: Nurhayati Br Tarigan, Rizkita Rachmi Esyanti

Abstract:

Stevia (Stevia rebaudiana Bertoni) has a great potential to be used as a natural sweetener because it contains steviol glycoside, which is approximately 100 - 300 times sweeter than sucrose, yet low calories. Vegetative and generative propagation of S. rebaudiana is inefficient to produce stevia biomass and stevioside. One of alternative for stevia propagation is in vitro shoot culture. This research was conducted to optimize the best medium for shoot growth and to compare the bioconversion efficiency and stevioside production of S. rebaudiana shoot culture cultivated in thin layer culture (TLC), recipient for automated temporary immersion system (TIS RITA®) bioreactor, and bubble column bioreactor. The result showed that 1 ppm of Kinetin produced a healthy shoot and the highest number of leaves compared to BAP. Shoots were then cultivated in TLC, TIS RITA® bioreactor, and bubble column bioreactor. Growth medium efficiency was determined by yield and productivity. TLC produced the highest growth medium efficiency of S. rebaudiana, the yield was 0.471 ± 0.117 gbiomass.gsubstrate-1, and the productivity was 0.599 ± 0.122 gbiomass.Lmedium-1.day-1. While TIS RITA® bioreactor produced the lowest yield and productivity, 0.182 ± 0.024 gbiomass.gsubstrate-1 and 0.041 ± 0.0002 gbiomass.Lmedium-1.day-1 respectively. The yield of bubble column bioreactor was 0.354 ± 0.204 gbiomass.gsubstrate-1 and the productivity was 0,099 ± 0,009 gbiomass.Lmedium-1.day-1. The stevioside content from the highest to the lowest was obtained from stevia shoot which was cultivated on TLC, TIS RITA® bioreactor, and bubble column bioreactor; the content was 93,44 μg/g, 42,57 μg/g, and 23,03 μg/g respectively. All three systems could be used to produce stevia shoot biomass, but optimization on the number of nutrition and oxygen intake was required in each system.

Keywords: bubble column, growth medium efficiency, Stevia rebaudiana, stevioside, TIS RITA®, TLC

Procedia PDF Downloads 239

Authors: Agustine Melviana, Rizkita Esyanti

Abstract:

Stevia plant contains steviol glycosides which is estimated to be 300 times sweeter than sucrose. However in Indonesia, conventional (in vivo) propagation of Stevia rebaudiana was not effective due to a poor result. Therefore, alternative methods to propagate S. rebaudiana plants is needed, one of it is using in vitro method. Multiplication with a large quantity of stevia biomass in relatively short period can be conducted by using TIS RITA® (Recipient for Automated Temporary Immersion System). The objective of this study was to evaluate the effect of immersion period of the medium on growth and the medium bioconversion into the production of shoot biomass. The study was conducted to determine the effect of different intensity period of medium to enhance biomass of stevia shoots. Shoot culture of S. rebaudiana was grown in full strength MS medium supplemented with 1 ppm Kinetin. RITA® bioreactors were set up with two different immersion periods, 15 min (RITA® 15) and 30 min (RITA® 30), scheduled every 6 hours and incubated for 21 days. The result indicated that immersion period affected the biomass and growth rate (µ). Thirty-minutes immersion showed greater percentage of shoot multiplication (93.44 ± 0.83%), percentage of leaf growth (85.24 ± 5.99%), growth rate (0.042 ± 0.001 g/day), and productivity (0.066 g/L medium/day) compared to that immersed in RITA® 15 min (76.90 ± 4.85%; 79.73 ± 7.76; 0.045 ± 0.004 g/day, and 0.045 g/L medium/day respectively). Enhancement of biomass in RITA® 30 reached 1,702 ± 0,114 gr, whereas in RITA® 15 only 0,953 ± 0,093 gr. Additionally, the pattern of sucrose, mineral, and inorganic compounds consumption followed the growth of plant biomass for both systems. In conclusion, the bioconversion efficiency from medium to biomass in RITA® 30 is better than RITA® 15.

Keywords: intensity period, shoot culture, Stevia rebaudiana, TIS RITA®

Procedia PDF Downloads 228

Authors: Asato Takaishi, Masashi Nasuhara, Ayuko Itsuki, Kenichi Suga

Abstract:

Stevia (Stevia rebaudiana Bertoni) is a composite plant native to Paraguay. Stevia sweetener is derived from a hot water extract of Stevia (Stevia extract), which has some effects such as histamine decomposition, antioxidative effect, and blood sugar level-lowering function. The steviol glycosides in the Stevia extract are considered to contribute to these effects. In addition, these effects increase by the fermentation. However, it takes a long time for fermentation of Stevia extract and the fermentation liquid sometimes decays during the fermentation process because natural fermentation method is used. The aim of this study is to perform the fermentation of Stevia extract in a shorter period, and to produce the fermentation liquid in stable quality. From the natural fermentation liquid of Stevia extract, the four strains of useful (good taste) microorganisms were isolated using dilution plate count method and some properties were determined. The base sequences of 16S rDNA and 28S rDNA revealed three bacteria (two Lactobacillus sp. and Microbacterium sp.) and one yeast (Issatchenkia sp.). This result has corresponded that several kinds of lactic bacterium such as Lactobacillus pentosus and Lactobacillus buchneri were isolated from Stevia leaves. Liquid chromatography/mass spectrometory (LC/MS/MS) and High-Performance Liquid Chromatography (HPLC) were used to determine the contents of steviol glycosides and neutral sugars. When these strains were cultured in the sterile Stevia extract, the steviol and stevioside were increased in the fermented Stevia extract. So, it was suggested that the rebaudioside A and the mixture of steviol glycosides in the Stevia extract were decomposed into stevioside and steviol by microbial metabolism.

Keywords: fermentation, lactobacillus, Stevia, steviol glycosides, yeast

Procedia PDF Downloads 516

Authors: Inayat Ur Rahman Arshad

Abstract:

Background: Stevia rebaudiana B. is a shrubby perennial herb of Asteraceae family that possesses the unique ability of accumulative non-caloric sweet steviol glycosides (SGs). Purpose: The purpose of the study is to optimize sugar concentration, pH level, and inoculum size for inducing the callus with optimum growth and efficient antibacterial potential. Method: Three different experiments were conducted in which Callus explant from three-months-old already established callus of Stevia reabudiana of four different sizes was inoculated on Murashige and Skoog (MS) basal medium supplemented with five different sucrose concentration and pH adjusted at four different levels. Results: Maximum callus induction 100, 87.5, and 85.33% resulted in the medium supplemented with 30 g/l sucrose, pH maintained at 5.5, and inoculated with 1.25g inoculum, respectively. Similarly, the highest fresh weights 65.00, 75.50, and 50.53 g/l were noted in a medium fortified with 40 g/l sucrose, inoculated 1.25g inoculum, and 6.0 pH level, respectively. However, the callus developed in a medium containing 50 g/l sucrose was found to be highly antibacterial potent with 27.3 and 26.5 mm inhibition zone against P. vulgaris and B. subtilis, respectively. Similarly, the callus grown on a medium inoculated with 1.00 g inoculum resulted in maximum antibacterial potential against S. aureus and P. vulgaris with 25 and 23.72 mm inhibition zone, respectively. However, in the case of pH levels, the medium maintained at 6.5 pH showed maximum antibacterial activity against P. vulgaris, B.subtilis, and E.coli with 27.9, 25, and 23.72 mm, respectively. The ethyl acetate extract of Stevia callus and leaves did not show antibacterial potential against Xanthomonas campestris and Clavebactor michiganensis. In the entire experiment, the standard antibacterial agent Streptomycin showed the highest inhibition zones among the rest of the callus extract; however, the pure dimethyl sulfoxide (DMSO) caused no inhibitory zone against any bacteria. Conclusion: From these findings, it is concluded that among various levels, sucrose @ 40 g L⁻¹, pH 6.0, and inoculums at 0.75 g were found best for most of the growth and quality attributes, including fresh weight, dry weight, and antibacterial activities and therefore can be recommended for callus proliferation and antibacterial potential of Stevia rebaudiana.

Keywords: Stevia rebaudiana, Steviol Glycosides, callus, Xanthomonas campestris

Procedia PDF Downloads 54

Authors: Umara Nissar Rafiqi, Irum Gul, Nazima Nasrullah, Monica Saifi, Malik Z. Abdin

Abstract:

Background: Stevia rebaudiana, a member of Asteraceae family is an important medicinal plant and produces a commercially used non-caloric natural sweetener, which is also an alternate herbal cure for diabetes. Steviol glycosides are the main sweetening compounds present in these plants. Secondary metabolites are crucial to the adaption of plants to the environment and its overcoming stress conditions. In agricultural procedures, the abiotic stresses like salinity, high metal toxicity and drought, in particular, are responsible for the majority of the reduction that differentiates yield potential from harvestable yield. Salt stress and heavy metal toxicity lead to increased production of reactive oxygen species (ROS). To avoid oxidative damage due to ROS and osmotic stress, plants have a system of anti-oxidant enzymes along with several stress induced enzymes. This helps in scavenging the ROS and relieve the osmotic stress in different cell compartments. However, whether stress induced toxicity modulates the activity of these enzymes in Stevia rebaudiana is poorly understood. Aim: The present study focussed on the effect of salinity, heavy metal toxicity (lead and mercury) on physiological traits and transcriptional profiling of Stevia rebaudiana. Method: Stevia rebaudiana plants were collected from the Central Institute of Medicinal and Aromatic plants (CIMAP), Patnagar, India and maintained under controlled conditions in a greenhouse at Hamdard University, Delhi, India. The plants were subjected to different concentrations of salt (0, 25, 50 and 75 mM respectively) and heavy metals, lead and mercury (0, 100, 200 and 300 µM respectively). The physiological traits such as shoot length, root numbers, leaf growth were evaluated. The samples were collected at different developmental stages and analysed for transcription profiling by RT-PCR. Transcriptional studies in stevia rebaudiana involves important antioxidant enzymes: catalase (CAT), superoxide dismutase (SOD), cytochrome P450 monooxygenase (CYP) and stress induced aquaporin (AQU), auxin repressed protein (ARP-1), Ndhc gene. The data was analysed using GraphPad Prism and expressed as mean ± SD. Result: Low salinity and lower metal toxicity did not affect the fresh weight of the plant. However, this was substantially decreased by 55% at high salinity and heavy metal treatment. With increasing salinity and heavy metal toxicity, the values of all studied physiological traits were significantly decreased. Chlorosis in treated plants was also observed which could be due to changes in Fe:Zn ratio. At low concentrations (upto 25 mM) of NaCl and heavy metals, we did not observe any significant difference in the gene expressions of treated plants compared to control plants. Interestingly, at high salt concentration and high metal toxicity, a significant increase in the expression profile of stress induced genes was observed in treated plants compared to control (p < 0.005). Conclusion: Stevia rebaudiana is tolerant to lower salt and heavy metal concentration. This study also suggests that with the increase in concentrations of salt and heavy metals, harvest yield of S. rebaudiana was hampered.

Keywords: Stevia rebaudiana, natural sweetener, salinity, heavy metal toxicity

Procedia PDF Downloads 169

Authors: Amir Ali, Laura Yael Mendoza

Abstract:

The application of nanomaterials is becoming the most effective strategy of elicitation to produce a desirable level of plant biomass with complex medicinal compounds. This study was designed to check the influence of phytosynthesized iron nanoparticles (FeNPs) on physical growth characteristics, antioxidant status, and production of steviol glycosides of in vitro grown Stevia rebaudiana. Effect of different concentrations of iron nanoparticles replacement of iron sulfate in MS medium (stock solution) on invitro stevia plant growth following positive control (MS basal medium), negative control (iron sulfate devoid medium), iron sulfate devoid MS medium and supplemented with FeNPs at different concentrations (5.6 mg/L, 11.2 mg/L, 16.8 mg/L, 22.4 mg/L) was evaluated. The iron deficiency leads to a drastic reduction in plant growth. In contrast, applying FeNPs leads to improvement in plant height, leave diameter, improved leave morphology, etc., in a concentration-dependent manner. Furthermore, the stress caused by FeNPs at 16.8 mg/L in cultures produced higher levels of total phenolic content (3.7 ± 0.042 mg/g dry weight: DW) and total flavonoid content (1.9 ± 0.022 mg/g DW and antioxidant activity (78 ± 4.6%). In addition, plants grown in the presence of FeNPs at 22.4 mg/L resulted in higher enzymatic antioxidant activities (SOD = 3.5 ± 0.042 U/mg; POD = 2.6 ± 0.026 U/mg; CAT = 2.8 ± 0.034 U/mg and APx = 3.6 ± 0.043 U/ mg), respectively. Furthermore, exposure to a higher dose of FeNPs (22.4 mg/L) exhibited the maximum amount of stevioside (stevioside: 4.6 ± 0.058 mg/g (DW) and rebaudioside A: 4.9 ± 0.068 mg/g DW) as compared to other doses. The current investigation confirms the effectiveness of FeNPs in growth media. It offers a suitable prospect for commercially desirable production of S. rebaudiana biomass with higher sweet glycosides profiles in vitro.

Keywords: cell culture, stevia, iron nanoparticles, antioxidants

Procedia PDF Downloads 67

Authors: Inayat Ur Rahman Arshad

Abstract:

Stevia rebaudiana B. is a shrubby perennial herb of Asteraceae family that possesses the unique ability of accumulative non caloric sweet Steviol Glycosides (SGs). The purpose of the study is to optimize sugar concentration, pH level and inoculum size for inducing the callus with optimum growth and efficient antibacterial potential. Three different experiments were conducted in which Callus explant from three-months-old already established callus of Stevia reabudiana of four different sizes were inoculated on Murashige and Skoog (MS) basal medium supplemented with five different sucrose concentration and pH adjusted at four different levels. Maximum callus induction 100, 87.5 and 85.33% was resulted in the medium supplemented with 30g/l sucrose, pH maintained at 5.5 and inoculated with 1.25g inoculum respectively. Similarly, the highest fresh weight 65.00, 75.50 and 50.53g/l were noted in medium fortified with 40g/l sucrose, inoculated 1.25g inoculum and 6.0 pH level respectively. However, the callus developed in medium containing 50g/l sucrose found highly antibacterial potent with 27.3 and 26.5mm inhibition zone against P. vulgaris and B. subtilize respectively. Similarly, the callus grown on medium inoculated with 1.00g inoculum resulted in maximum antibacterial potential against S. aureus and P. vulgaris with 25 and 23.72mm inhibition zones respectively. However, in the case of pH levels the medium maintained at 6.5pH showed maximum antibacterial activity against P. vulgaris, B.subtilis and E.coli with 27.9, 25 and 23.72mm respectively. The ethyl acetate extract of Stevia callus and leaves did not show antibacterial potential against Xanthomonas campestris and Clavebactor michiganensis. In the entire experiment the standard antibacterial agent Streptomycin showed the highest inhibition zones from the rest of the callus extract, however the pure DMSO (Dimethyl Sulfoxide) caused no inhibitory zone against any bacteria. From these findings it is concluded that among various levels sucrose at the rate of 40g L-1, pH 6.0 and inoculums 0.75g was found best for most of the growth and quality attributes including fresh weight, dry weight and antibacterial activities and therefore can be recommended for callus proliferation and antibacterial potential of Stevia rebaudiana

Keywords: Steviol Glycosides, Skoog, Murashige, Clavebactor michiganensis

Procedia PDF Downloads 60

Authors: Rafael A. B. De Medeiros, Zilmar M. P. Barros, Carlos B. O. De Carvalho, Eunice G. Fraga Neta, Maria I. S. Maciel, Patricia M. Azoubel

Abstract:

The use of the sugar substitution technique (D3S) in mango was studied. It consisted of two stages and the use of ultrasound in one or both stages was evaluated in terms of water loss and solid gain. Higher water loss results were found subjecting the fruit samples to ultrasound in the first stage followed by immersion of the samples in Stevia-based solution with application of ultrasound in the second stage, while higher solids gain were obtained without application of ultrasound in second stage. Samples were evaluated in terms of total carotenoids content and total color difference. Samples submitted to ultrasound in both D3S stages presented higher carotenoid retention compared to samples sonicated only in the first stage. Color of man goes after the D3S process showed notable changes.

Keywords: Mangifera indica L., quality, Stevia rebaudiana, ultrasound

Procedia PDF Downloads 369

Authors: Li Xiaoge, Wang Chaodong

Abstract:

We introduce the first Chinese Coreference Resolution Model based on BERT (CCRM-BERT) and show that it significantly outperforms all previous work. The key idea is to consider the features of the mention, such as part of speech, width of spans, distance between spans, etc. And the influence of each features on the model is analyzed. The model computes mention embeddings that combine BERT with features. Compared to the existing state-of-the-art span-ranking approach, our model significantly improves accuracy on the Chinese OntoNotes benchmark.

Keywords: BERT, coreference resolution, deep learning, nature language processing

Procedia PDF Downloads 172

Authors: Qitao Xie, Qingquan Zhang, Xiaofei Zhang, Di Tian, Ruixuan Wen, Ting Zhu, Ping Yi, Xin Li

Abstract:

Inspired by the recent movement of digital currency, we are building a question answering system concerning the subject of cryptocurrency using Bidirectional Encoder Representations from Transformers (BERT). The motivation behind this work is to properly assist digital currency investors by directing them to the corresponding knowledge bases that can offer them help and increase the querying speed. BERT, one of newest language models in natural language processing, was investigated to improve the quality of generated responses. We studied different combinations of hyperparameters of the BERT model to obtain the best fit responses. Further, we created an intelligent chatbot for cryptocurrency using BERT. A chatbot using BERT shows great potential for the further advancement of a cryptocurrency market tool. We show that the BERT neural networks generalize well to other tasks by applying it successfully to cryptocurrency.

Keywords: bidirectional encoder representations from transformers, BERT, chatbot, cryptocurrency, deep learning

Procedia PDF Downloads 108

Authors: Gurveer Kaur, T. K. Goswami

Abstract:

Rasgulla (a sweet syrup cheese ball), a sweet, spongy dessert represents traditional sweet dish of an Indian subcontinent prepared by chhana. 100 g of Rasgulla contains 186 calories, and so it is a driving force behind obesity and diabetes. To reduce Rasgulla’s energy value sucrose mainly should be minimized, so instead of sucrose, stevia (zero calories natural sweetener) is used to prepare Rasgulla. In this study three samples were prepared with sucrose to stevia ratio taking 100:0 (as control sample), (i) 50:50 (T1); (ii) 25:75 (T2), and (iii) 0:100 (T3) from 4% fat milk. It was found that as the sucrose concentration decreases the percentage of fat increase in the Rasgulla slightly. Sample T2 showed < 0.1% (±0.06) sucrose content. But there was no significant difference on protein and ash content of the samples. Whitening index was highest (78.0 ± 0.13) for T2 and lowest (65.7 ± 0.21) for the control sample since less sucrose in syrup reduces the browning of the sample (T2). Energy value per 100 g was calculated to be 50, 72, 98, and 184 calories for T3, T2, T1 and control samples, respectively. According to optimization study, the preferred (high quality) order of samples was as follows: T1 > T1 > control > T3. Low sugar content Rasgulla with acceptable quality can be prepared with 25:75 ratio of sucrose to stevia.

Keywords: composition, rasgulla, sensory, stevia

Procedia PDF Downloads 178

Authors: Sayyed-Javad Ziaolhagh, Mojtaba Hokmabadi

Abstract:

Introduction: Childhood obesity is a serious medical condition that affects children and adolescents even in the central nervous system. Semaphorins also play a role in the inflammatory process of the nervous system. On the other hand, it has been stated that garlic and stevia extracts following aerobic exercise are effective on immune system inflammation in addition to aerobic activity. Materials and Methods: For 15 weeks, 50 3-week-old male Wistar rats were fed with conventional rodent chow for control and a high-fat diet to induce obesity. Obese rats then were randomly assigned into 7 groups (n=5) based on the Lee index: healthy control (C), obese (OBS), obese + garlic (OBS+GAR), obese + Stevia (OBS+STV), obese + aerobic exercise (OBS+EXE), obese + garlic + aerobic exercise (OBS+GAR+EXE), and obese + stevia + aerobic exercise (OBS+STV+EXE). Training groups completed a progressive aerobic running program (at 8-15 m/min, 5-20 min/day, 5 days/week), and Stevia and garlic extract group (250 mg/kg/day, 5 days/week) were given orally once a day. Real-time PCR was used to determine the levels of Semaphorin 4A, and Plexin D1 gene expressions in the hypothalamus. Fold change analysis with ANOVA was performed for statistical analysis, with a significance threshold of P<0.05. Results: Body weight increased significantly in OBS compared to C (p= 0.013), but was not significantly changed in all treatment rats. Moreover, Semaphorin 4A was significantly increased in obese compared to control group (p= 0.041) and after 8 weeks, stevia extract (p=0.006), aerobic exercise (p=0.012) and garlic extract + aerobic exercise (p=0.008) significantly decreased compared to obese rats. In addition, Plexin D1 genes were also found in the hypothalamus of both obese and control rats but were insignificantly up-regulated when compared with the obese group (p=0.950). Conclusion: High-fat diet caused neuroinflammation by elevation of sema4A in obese rats and stevia, stevia with aerobic and garlic with aerobic could reduce this inflammation in rats. Also, none of them could alter Plexin D1.

Keywords: sema 4A, plexin D1, garlic, stevia

Procedia PDF Downloads 47

Authors: Dylan Elliott, Katya Pertsova

Abstract:

Preposition errors are some of the most common errors created by L2 speakers. In addition, improving error correction and detection methods remains an open issue in the realm of Natural Language Processing (NLP). This research investigates whether the bidirectional encoder representations from the transformers model (BERT) have the potential to correct preposition errors accurately enough to be useful in error correction software. This research finds that BERT performs strongly when the scope of its error correction is limited to preposition choice. The researchers used an open-source BERT model and over three hundred thousand edited sentences from Wikipedia, tagged for part of speech, where only a preposition edit had occurred. To test BERT’s ability to detect errors, a technique known as multi-level masking was used to generate suggestions based on sentence context for every prepositional environment in the test data. These suggestions were compared with the original errors in the data and their known corrections to evaluate BERT’s performance. The suggestions were further analyzed to determine if BERT more often agreed with the judgements of the Wikipedia editors. Both the untrained and fined-tuned models were compared. Finetuning led to a greater rate of error-detection which significantly improved recall, but lowered precision due to an increase in false positives or falsely flagged errors. However, in most cases, these false positives were not errors in preposition usage but merely cases where more than one preposition was possible. Furthermore, when BERT correctly identified an error, the model largely agreed with the Wikipedia editors, suggesting that BERT’s ability to detect misused prepositions is better than previously believed. To evaluate to what extent BERT’s false positives were grammatical suggestions, we plan to do a further crowd-sourcing study to test the grammaticality of BERT’s suggested sentence corrections against native speakers’ judgments.

Keywords: BERT, grammatical error correction, preposition error detection, prepositions

Procedia PDF Downloads 116

Authors: Amnah M. A. Alsuhaibani, Amal N. Al-Kuraieef

Abstract:

Introduction: In the recent decades, production of low-calorie jams is needed for diabetics that comprise low calorie fruits and low calorie sweeteners. Object: the research aimed to prepare low calorie formulated pumpkin jams (fructose, stevia and aspartame) incorporated with soy bean and evaluate the jams through chemical analysis and sensory evaluation after storage for six month. Moreover, the possible effect of consumption of low calorie jams on diabetic rats was investigated. Methods: Five formulas of pumpkin jam with different sucrose, fructose, stevia and aspartame sweeteners and soy bean were prepared and stored at 10 oC for six month compared to ordinary pumpkin jam. Chemical composition and sensory evaluation of formulated jams were evaluated at zero time, 3 month and 6 month of storage. The best three acceptable pumpkin jams were taken for biological study on diabetic rats. Rats divided into group (1) served as negative control and streptozotocin induce diabetes four rat groups that were positive diabetic control (group2), rats fed on standard diet with 10% sucrose soybean jam, fructose soybean jam and stevia soybean jam (group 3, 4&5), respectively. Results: The content of protein, fat, ash and fiber were increased but carbohydrate was decreased in low calorie formulated pumpkin jams compared to ordinary jam. Production of aspartame soybean pumpkin jam had lower score of all sensory attributes compared to other jam then followed by stevia soybean Pumpkin jam. Using non nutritive sweeteners (stevia & aspartame) with soybean in processing jam could lower the score of the sensory attributes after storage for 3 and 6 months. The highest score was recorded for sucrose and fructose soybean jams followed by stevia soybean jam while aspartame soybean jam recorded the lowest score significantly. The biological evaluation showed a significant improvement in body weight and FER of rats after six weeks of consumption of standard diet with jams (Group 3,4&5) compared to Group1. Rats consumed 10% low calorie jam with nutrient sweetener (fructose) and non nutrient sweetener (stevia) soybean jam (group 4& 5) showed significant decrease in glucose level, liver function enzymes activity, and liver cholesterol & total lipids in addition of significant increase of insulin and glycogen compared to the levels of group 2. Conclusion: low calorie pumpkin jams can be prepared by low calorie sweeteners and soybean and also storage for 3 months at 10oC without change sensory attributes. Consumption of stevia pumpkin jam fortified with soybean had positive health effects on streptozoticin induced diabetes in rats.

Keywords: pumpkin jam, HFCS, aspartame, stevia, storage

Procedia PDF Downloads 152

Authors: Adeep Hande, Shubham Agarwal

Abstract:

This paper presents a study on identifying sexism in online texts using various state-of-the-art deep learning models based on BERT. We experimented with different feature sets and model architectures and evaluated their performance using precision, recall, F1 score, and accuracy metrics. We also explored the use of pseudolabeling technique to improve model performance. Our experiments show that the best-performing models were based on BERT, and their multilingual model achieved an F1 score of 0.83. Furthermore, the use of pseudolabeling significantly improved the performance of the BERT-based models, with the best results achieved using the pseudolabeling technique. Our findings suggest that BERT-based models with pseudolabeling hold great promise for identifying sexism in online texts with high accuracy.

Keywords: large language models, semi-supervised learning, sexism detection, data sparsity

Procedia PDF Downloads 36

Authors: Esther Chiramal, Kelvin Soh Boon Kai

Abstract:

Explainable AI is a strong strategy implemented to understand complex black-box model predictions in a human-interpretable language. It provides the evidence required to execute the use of trustworthy and reliable AI systems. On the other hand, however, it also opens the door to locating possible vulnerabilities in an AI model. Traditional adversarial text attack uses word substitution, data augmentation techniques, and gradient-based attacks on powerful pre-trained Bidirectional Encoder Representations from Transformers (BERT) variants to generate adversarial sentences. These attacks are generally white-box in nature and not practical as they can be easily detected by humans e.g., Changing the word from “Poor” to “Rich”. We proposed a simple yet effective Grey-box cum Black-box approach that does not require the knowledge of the model while using a set of surrogate Transformer/BERT models to perform the attack using Explainable AI techniques. As Transformers are the current state-of-the-art models for almost all Natural Language Processing (NLP) tasks, an attack generated from BERT1 is transferable to BERT2. This transferability is made possible due to the attention mechanism in the transformer that allows the model to capture long-range dependencies in a sequence. Using the power of BERT generalisation via attention, we attempt to exploit how transformers learn by attacking a few surrogate transformer variants which are all based on a different architecture. We demonstrate that this approach is highly effective to generate semantically good sentences by changing as little as one word that is not detectable by humans while still fooling other BERT models.

Keywords: BERT, explainable AI, Grey-box text attack, transformer

Procedia PDF Downloads 112

Authors: Hsin-Yang Wang, K. M. A. Salam, Ying-Jia Lin, Daniel Tan, Tzu-Hsuan Chou, Hung-Yu Kao

Abstract:

Detecting user intent from natural language expression has a wide variety of use cases in different natural language processing applications. Recently few-shot training has a spike of usage on commercial domains. Due to the lack of significant sample features, the downstream task performance has been limited or leads to an unstable result across different domains. As a state-of-the-art method, the pre-trained BERT model gathering the sentence-level information from a large text corpus shows improvement on several NLP benchmarks. In this research, we are proposing a method to change multi-class classification tasks into binary classification tasks, then use the confidence score to rank the results. As a language model, BERT performs well on sequence data. In our experiment, we change the objective from predicting labels into finding the relations between words in sequence data. Our proposed method achieved 71.0% accuracy in the internal intent detection dataset and 63.9% accuracy in the HuffPost dataset. Acknowledgment: This work was supported by NCKU-B109-K003, which is the collaboration between National Cheng Kung University, Taiwan, and SoftBank Corp., Tokyo.

Keywords: OSML, BERT, text classification, one shot

Procedia PDF Downloads 79

Authors: AnjalI A. Prasad

Abstract:

Nowadays, social media played an important role in spreading misinformation or fake news. This study analyzes the fake news related to the COVID-19 pandemic spread in social media. This paper aims at evaluating and comparing different approaches that are used to mitigate this issue, including popular deep learning approaches, such as CNN, RNN, LSTM, and BERT algorithm for classification. To evaluate models’ performance, we used accuracy, precision, recall, and F1-score as the evaluation metrics. And finally, compare which algorithm shows better result among the four algorithms.

Keywords: BERT, CNN, LSTM, RNN

Procedia PDF Downloads 176

Authors: Moquan Gong

Abstract:

In recent years, with the rapid development of deep learning, pre-training technology for natural language processing has made great progress. The early field of natural language processing has long used word vector methods such as Word2Vec to encode text. These word vector methods can also be regarded as static pre-training techniques. However, this context-free text representation brings very limited improvement to subsequent natural language processing tasks and cannot solve the problem of word polysemy. ELMo proposes a context-sensitive text representation method that can effectively handle polysemy problems. Since then, pre-training language models such as GPT and BERT have been proposed one after another. Among them, the BERT model has significantly improved its performance on many typical downstream tasks, greatly promoting the technological development in the field of natural language processing, and has since entered the field of natural language processing. The era of dynamic pre-training technology. Since then, a large number of pre-trained language models based on BERT and XLNet have continued to emerge, and pre-training technology has become an indispensable mainstream technology in the field of natural language processing. This article first gives an overview of pre-training technology and its development history, and introduces in detail the classic pre-training technology in the field of natural language processing, including early static pre-training technology and classic dynamic pre-training technology; and then briefly sorts out a series of enlightening technologies. Pre-training technology, including improved models based on BERT and XLNet; on this basis, analyze the problems faced by current pre-training technology research; finally, look forward to the future development trend of pre-training technology.

Keywords: natural language processing, pre-training, language model, word vectors

Procedia PDF Downloads 22

Authors: Félix David Suárez Bonilla

Abstract:

A sentiment analysis service to detect polarity (positive, neural, and negative), based on transfer learning, was built using a Spanish version of BERT and applied to tweets written in Spanish. The dataset that was used consisted of 11975 reviews, which were extracted from Google Play using the google-play-scrapper package. The BETO trained model used: the AdamW optimizer, a batch size of 16, a learning rate of 2x10⁻⁵ and 10 epochs. The system was tested using tweets of three presidential pre-candidates from Costa Rica. The system was finally validated using human labeled examples, achieving an accuracy of 83.3%.

Keywords: NLP, transfer learning, BERT, sentiment analysis, social media, opinion mining

Procedia PDF Downloads 139

Authors: Jamshid Qodirov, Sirojiddin Komolov, Ravilov Mirahmad, Olimjon Mirzayev

Abstract:

Nowadays, natural language processing tools play a crucial role in our daily lives, including various techniques with text processing. There are very advanced models in modern languages, such as English, Russian etc. But, in some languages, such as Uzbek, the NLP models have been developed recently. Thus, there are only a few NLP models in Uzbek language. Moreover, there is no such work that could show which Uzbek NLP model behaves in different situations and when to use them. This work tries to close this gap and compares the Uzbek NLP models existing as of the time this article was written. The authors try to compare the NLP models in two different scenarios: sentiment analysis and sentence similarity, which are the implementations of the two most common problems in the industry: classification and similarity. Another outcome from this work is two datasets for classification and sentence similarity in Uzbek language that we generated ourselves and can be useful in both industry and academia as well.

Keywords: NLP, benchmak, bert, vectorization

Procedia PDF Downloads 25

Authors: Josiel Delgadillo, Johnson Kinyua, Charles Mutigwe

Abstract:

The purpose of sentiment analysis is to determine the sentiment strength (e.g., positive, negative, neutral) from a textual source for good decision-making. Natural language processing in domains such as financial markets requires knowledge of domain ontology, and pre-trained language models, such as BERT, have made significant breakthroughs in various NLP tasks by training on large-scale un-labeled generic corpora such as Wikipedia. However, sentiment analysis is a strong domain-dependent task. The rapid growth of social media has given users a platform to share their experiences and views about products, services, and processes, including financial markets. StockTwits and Twitter are social networks that allow the public to express their sentiments in real time. Hence, leveraging the success of unsupervised pre-training and a large amount of financial text available on social media platforms could potentially benefit a wide range of financial applications. This work is focused on sentiment analysis using social media text on platforms such as StockTwits and Twitter. To meet this need, SkyBERT, a domain-specific language model pre-trained and fine-tuned on financial corpora, has been developed. The results show that SkyBERT outperforms current state-of-the-art models in financial sentiment analysis. Extensive experimental results demonstrate the effectiveness and robustness of SkyBERT.

Keywords: BERT, financial markets, Twitter, sentiment analysis

Procedia PDF Downloads 123

Authors: Ebipatei Victoria Tunyan, T. A. Cao, Cheol Young Ock

Abstract:

Detecting subjectively biased statements is a vital task. This is because this kind of bias, when present in the text or other forms of information dissemination media such as news, social media, scientific texts, and encyclopedias, can weaken trust in the information and stir conflicts amongst consumers. Subjective bias detection is also critical for many Natural Language Processing (NLP) tasks like sentiment analysis, opinion identification, and bias neutralization. Having a system that can adequately detect subjectivity in text will boost research in the above-mentioned areas significantly. It can also come in handy for platforms like Wikipedia, where the use of neutral language is of importance. The goal of this work is to identify the subjectively biased language in text on a sentence level. With machine learning, we can solve complex AI problems, making it a good fit for the problem of subjective bias detection. A key step in this approach is to train a classifier based on BERT (Bidirectional Encoder Representations from Transformers) as upstream model. BERT by itself can be used as a classifier; however, in this study, we use BERT as data preprocessor as well as an embedding generator for a Bi-LSTM (Bidirectional Long Short-Term Memory) network incorporated with attention mechanism. This approach produces a deeper and better classifier. We evaluate the effectiveness of our model using the Wiki Neutrality Corpus (WNC), which was compiled from Wikipedia edits that removed various biased instances from sentences as a benchmark dataset, with which we also compare our model to existing approaches. Experimental analysis indicates an improved performance, as our model achieved state-of-the-art accuracy in detecting subjective bias. This study focuses on the English language, but the model can be fine-tuned to accommodate other languages.

Keywords: subjective bias detection, machine learning, BERT–BiLSTM–Attention, text classification, natural language processing

Procedia PDF Downloads 99

Authors: Melissa Vang, Raheyma Khan, Haihua Chen

Abstract:

The coronavirus disease 2019 (COVID‐19) pandemic has changed people's lives from all over the world. More people have turned to Twitter to engage online and discuss the COVID-19 vaccine. This study aims to present a text mining approach to identify people's attitudes towards the COVID-19 vaccine on Twitter. To achieve this purpose, we collected 54,268 COVID-19 vaccine tweets from September 01, 2020, to November 01, 2020, then the BERT model is used for the sentiment and topic analysis. The results show that people had more negative than positive attitudes about the vaccine, and countries with an increasing number of confirmed cases had a higher percentage of negative attitudes. Additionally, the topics discussed in positive and negative tweets are different. The tweet datasets can be helpful to information professionals to inform the public about vaccine-related informational resources. Our findings may have implications for understanding people's cognitions and feelings about the vaccine.

Keywords: BERT, COVID-19 vaccine, sentiment analysis, topic modeling

Procedia PDF Downloads 118

Authors: Burak Oksuzoglu, Savas Yildirim, Ferhat Kutlu

Abstract:

Opinion mining (OM) is one of the natural language processing (NLP) problems to determine the polarity of opinions, mostly represented on a positive-neutral-negative axis. The data for OM is usually collected from various social media platforms. In an era where social media has considerable control over companies’ futures, it’s worth understanding social media and taking actions accordingly. OM comes to the fore here as the scale of the discussion about companies increases, and it becomes unfeasible to gauge opinion on individual levels. Thus, the companies opt to automize this process by applying machine learning (ML) approaches to their data. For the last two decades, OM or sentiment analysis (SA) has been mainly performed by applying ML classification algorithms such as support vector machines (SVM) and Naïve Bayes to a bag of n-gram representations of textual data. With the advent of deep learning and its apparent success in NLP, traditional methods have become obsolete. Transfer learning paradigm that has been commonly used in computer vision (CV) problems started to shape NLP approaches and language models (LM) lately. This gave a sudden rise to the usage of the pretrained language model (PTM), which contains language representations that are obtained by training it on the large datasets using self-supervised learning objectives. The PTMs are further fine-tuned by a specialized downstream task dataset to produce efficient models for various NLP tasks such as OM, NER (Named-Entity Recognition), Question Answering (QA), and so forth. In this study, the traditional and modern NLP approaches have been evaluated for OM by using a sizable corpus belonging to a large private company containing about 76,000 comments in Turkish: SVM with a bag of n-grams, and two chosen pre-trained models, multilingual universal sentence encoder (MUSE) and bidirectional encoder representations from transformers (BERT). The MUSE model is a multilingual model that supports 16 languages, including Turkish, and it is based on convolutional neural networks. The BERT is a monolingual model in our case and transformers-based neural networks. It uses a masked language model and next sentence prediction tasks that allow the bidirectional training of the transformers. During the training phase of the architecture, pre-processing operations such as morphological parsing, stemming, and spelling correction was not used since the experiments showed that their contribution to the model performance was found insignificant even though Turkish is a highly agglutinative and inflective language. The results show that usage of deep learning methods with pre-trained models and fine-tuning achieve about 11% improvement over SVM for OM. The BERT model achieved around 94% prediction accuracy while the MUSE model achieved around 88% and SVM did around 83%. The MUSE multilingual model shows better results than SVM, but it still performs worse than the monolingual BERT model.

Keywords: BERT, MUSE, opinion mining, pretrained language model, SVM, Turkish

Procedia PDF Downloads 112

Authors: Kuo-Chun Chien, Chia-Hui Chang, Ren-Der Sun

Abstract:

Legal Judgment Prediction (LJP) is a subtask for legal AI. Its main purpose is to use the facts of a case to predict the judgment result. In Taiwan's criminal procedure, when prosecutors complete the investigation of the case, they will decide whether to prosecute the suspect and which article of criminal law should be used based on the facts and evidence of the case. In this study, we collected 305,240 indictments from the public inquiry system of the procuratorate of the Ministry of Justice, which included 169 charges and 317 articles from 21 laws. We take the crime facts in the indictments as the main input to jointly learn the prediction model for law source, article, and charge simultaneously based on the pre-trained Bert model. For single article cases where the frequency of the charge and article are greater than 50, the prediction performance of law sources, articles, and charges reach 97.66, 92.22, and 60.52 macro-f1, respectively. To understand the big performance gap between articles and charges, we used a bipartite graph to visualize the relationship between the articles and charges, and found that the reason for the poor prediction performance was actually due to the wording precision. Some charges use the simplest words, while others may include the perpetrator or the result to make the charges more specific. For example, Article 284 of the Criminal Law may be indicted as “negligent injury”, "negligent death”, "business injury", "driving business injury", or "non-driving business injury". As another example, Article 10 of the Drug Hazard Control Regulations can be charged as “Drug Control Regulations” or “Drug Hazard Control Regulations”. In order to solve the above problems and more accurately predict the article and charge, we plan to include the article content or charge names in the input, and use the sentence-pair classification method for question-answer problems in the BERT model to improve the performance. We will also consider a sequence-to-sequence approach to charge prediction.

Keywords: legal judgment prediction, deep learning, natural language processing, BERT, data visualization

Procedia PDF Downloads 96

Authors: A. M. John-Otumu, M. M. Rahman, O. C. Nwokonkwo, M. C. Onuoha

Abstract:

Online social media networks have long served as a primary arena for group conversations, gossip, text-based information sharing and distribution. The use of natural language processing techniques for text classification and unbiased decision-making has not been far-fetched. Proper classification of this textual information in a given context has also been very difficult. As a result, we decided to conduct a systematic review of previous literature on sentiment classification and AI-based techniques that have been used in order to gain a better understanding of the process of designing and developing a robust and more accurate sentiment classifier that can correctly classify social media textual information of a given context between hate speech and inverted compliments with a high level of accuracy by assessing different artificial intelligence techniques. We evaluated over 250 articles from digital sources like ScienceDirect, ACM, Google Scholar, and IEEE Xplore and whittled down the number of research to 31. Findings revealed that Deep learning approaches such as CNN, RNN, BERT, and LSTM outperformed various machine learning techniques in terms of performance accuracy. A large dataset is also necessary for developing a robust sentiment classifier and can be obtained from places like Twitter, movie reviews, Kaggle, SST, and SemEval Task4. Hybrid Deep Learning techniques like CNN+LSTM, CNN+GRU, CNN+BERT outperformed single Deep Learning techniques and machine learning techniques. Python programming language outperformed Java programming language in terms of sentiment analyzer development due to its simplicity and AI-based library functionalities. Based on some of the important findings from this study, we made a recommendation for future research.

Keywords: artificial intelligence, natural language processing, sentiment analysis, social network, text

Procedia PDF Downloads 91

Authors: Erekle Magradze, Davit Soselia, Levan Shughliashvili, Irakli Koberidze, Shota Tsiskaridze, Victor Kakhniashvili, Tamar Chaghiashvili

Abstract:

Since the beginning of active publishing activity in Georgia, voluminous printed material has been accumulated, the digitization of which is an important task. Digitized materials will be available to the audience, and it will be possible to find text in them and conduct various factual research. Digitizing scanned documents means scanning documents, extracting text from the scanned documents, and processing the text into a corresponding language model to detect inaccuracies and grammatical errors. Implementing these stages requires a unified, scalable, and automated platform, where the digital service developed for each stage will perform the task assigned to it; at the same time, it will be possible to develop these services dynamically so that there is no interruption in the work of the platform.

Keywords: NLP, OCR, BERT, Kubernetes, transformers

Procedia PDF Downloads 114