34254 research outputs found
Sort by
A Study on the Relationship among Accouting Staff\ue2s Personality Traits, Role Stress and Internal Auditing Qualities-The Case in Penghu County
No full textThe accounting staff are dispatched to various organizations by their respective accounting agencies, empowered by laws and regulations to independently carry out internal auditing functions. The internal auditing qualities not only significantly affects the financial efficiency of these organizations but also effectively helps them avoid external audit deficiencies and accountability. However, during the process of internal auditing, the accounting staff often encounter doubts with service units regarding write-offs, leading to challenges in the execution of internal audits over time.
The aim of this study is to understand the perceived scenarios of personality traits, role stress, and internal auditing qualities, as well as to explore their interrelationships. The research participants of this study are the accounting staff of public agencies in Penghu County. This study employed a questionnaire survey, which was conducted using a census method. Out of 93 questionnaires collected, 92 were valid, yielding an effective response rate of 98.9%. Data from the returned samples were processed and analyzed using the SPSS statistical software 28.0, using quantitative statistical methods to verify the research hypotheses. The results are as follows:
1. The most prominent personality trait of the accounting staff was their conscientiousness. Their identification with role stress was moderately low, while their perception of the qualities of internal auditing showed a moderately high identification.
2. For personality traits, role stress, and internal auditing qualities, individual background variables exhibited significant differences in certain aspects.
3. A significant negative correlation was observed between personality traits and role stress.
4. A significant negative correlation was observed between role stress and internal auditing qualities.
5. A significant positive correlation was observed between personality traits and internal auditing qualities.
This study summarizes the empirical results and makes specific recommendations for the future management of accountants, business units, and authoritie
Relationship analysis between the vinyl chloride emission in stacks and perimeter concentrations by Gaussian Plume Model
No full textThe Environmental Protection Agency of the Executive Yuan announced the emission standards of harmful air substances from fixed pollution sources on February 26, 2021, giving priority to the control of 72 important harmful air pollutants and setting the standards for discharge pipes and perimeters, among which vinyl chloride belongs to the first category Human carcinogens have a greater impact on human health, so the emission standards have been tightened, and different implementation times have been set for existing pollution sources in two stages. The first stage perimeter standard of 40 ppb has been implemented on July 1, 2021 , the second phase of the perimeter standard is 20 ppb on July 1, 2023.
In order to investigate whether the impact of the current VCM emissions on the perimeter of Plant A complies with the tightened perimeter standards, this study uses the Gaussian diffusion model to assess whether the concentration of vinyl chloride at the perimeter of Plant A will exceed the regulations under different atmospheric conditions. Tighten standards. In this study, the three models of Gaussian diffusion are first compared, and the simulation parameters are set based on the maximum emission source and site data of plant A, and then substituted into the basic model, mirror model, and inversion layer model for simulation.
The results show that the maximum concentration of pollutants diffused in the perimeter from high to low is the inversion layer model, the mirror model, and the basic model, and the perimeter concentrations are 38.16, 12.39, and 6.20 \uce\ubcg/m3, respectively. The inversion layer model considers many factors and is closer to the actual diffusion situation, so it is used as a model for subsequent evaluation. After determining the model, discuss the diffusion factors (wind speed and atmospheric stability) that affect the concentration of pollutants. The perimeter of plant A is about 30-600 m away from the emission source. Under the conditions of wind speed 1, 2 and 3 m/s, within the range of 30-600 m around the perimeter, the concentration results show that the maximum concentration of pollutants diffuses from high to low The wind speeds are 1, 2, and 3 m/s in sequence, respectively 71.69, 35.85, and 23.90 \uce\ubcg/m3, indicating that when the wind speed is smaller, the pollutants are less likely to diffuse and the diffusion concentration is higher. lower. Under different atmospheric stability conditions, when the atmospheric stability is B (moderately unstable) and C (slightly unstable), the concentration of pollutant diffusion is higher, and the distance from the emission source is closer, with the maximum values of 35.85 and 38.16 \uce\ubcg/m3 respectively , which are located at 170 and 250 m downwind, respectively; when the atmospheric stability is E (slightly stable) and F (moderately stable), the pollutant diffusion concentration is low and the distance from the emission source is far away, and the maximum values are 29.80 and 10.73 \uce\ubcg/m3, at a distance of 600 m from the emission source. Overall, when the atmospheric stability is C, the concentration at the perimeter will not exceed the tightened standard; when the atmospheric stability is B, the wind speed will exceed the tightened standard when the wind speed is below 1.4 m/s.
Finally, the perimeter tightening standard is set as the diffusion concentration after the simulation, and the emission volume of the pipeline that meets the perimeter tightening standard is calculated back with different atmospheric diffusion factors. The results show that when the wind speed decreases, the pollutants are not easy to spread, and the emission must be reduced to meet the tightened perimeter standard. When the wind speed is 1 m/s, the emission must be lower than 0.026 g/s to meet the perimeter tightened standard, but According to the average emissions of the 9 discharge pipes of Plant A from 2020 to 2022 were 0.005, 0.004, and 0.004 g/s respectively, it shows that the current emissions can still meet the tightened perimeter concentration standards
Radar System using Self-Injection-Locked Technology
No full textThis thesis presents the application of self-injection locking technology in radar system by proposing a heterodyne self-injection-locked radar (HSILR). Compared to traditional radar systems, HSILR offers improved sensitivity, reduced clutter and environmental interference, and lower costs. The sensitivity of the radar system is significantly affected by the self-injection-locked oscillator (SILO)'s operating frequency. Directly lowering the center frequency of SILO would result in a decrease in radar system performance. Therefore, this thesis employs a frequency converter to lower the transmission frequency from 920 MHz to 30 MHz, allowing the utilization of high-frequency SILO's sensitivity to enhance the system's signal-to-noise ratio (SNR). The received signal is also up-converted to avoid instability caused by a large difference between the injection signal and the original oscillation frequency.
The radar system architecture presented in this thesis posesses significant advantages in terms of frequency planning. It enables the operation of main RF components in commonly used commercial frequency bands, which facilitates the use of off-the-shelf products to accelerate development time. The use of a Direct Digital Synthesizer for frequency modulation effectively eliminates the possibility of ghost targets caused by nonlinear scanning and allows for flexible adjustments of the scanning bandwidth. The proposed radar system underwent initial validation of its components and sub-systems, and finally, it was tested in both Continuous Wave (CW) and Frequency Modulated Continuous Wave (FMCW) modes to verify its capabilities in obtaining accurate Doppler information and range measurement
Implementation of Ring Oscillators with Variation Suppression Based On Optimized Layout Analysis of Delay Elements
No full textWith the fast advance of the CMOS processes in these years, the impact caused by various variations have been seriously considered in chip design and fabrication. The major important variations are Process, Voltage, Temperature, and Leakage (PVTL). Notably, the last one is the most significant in the advanced processes. To investigate whether there is any variation caused by the layout given the same circuit design, this thesis proposes to take advantage of various delay component arrangements of generic ring oscillators and realize them using a typical CMOS process to find out the relationship between layouts and variation reduction.
In Chapter 2, the said ring oscillator is composed of dual-input dual-output delay cells with a multiplexer and a decoder to constitute a digital signal-controlled ring oscillator (RO). Namely, different delay cell arrangements are realized to find a way to reduce the layout variation. The frequency of RO can be changed using a 6-bit digital code. The on-silicon measurement results showed that all chips could work, but the output frequency was not matched with the simulation counterparts.
In Chapter 3, when the RO is affected by variations (PVT), the PVT variation detection circuit (including Process, Voltage, and Temperature Variation Detector) can detect the corner and calibrate the output frequency close to the expected frequency. Therefore, if the frequency of the oscillator simulation is as same as the measured frequency on chip, the frequency of the oscillator simulation at 45 different corners is compiled into a table, and generalized the method to adjust the frequency back to the expected frequency
Exploring the impact of geographic location tracking surveillance on romantic relationships
No full textNowadays, couples have turned to CMC to gather more information about their significant others. One prevalent approach is the utilization of location-based social networking (LBSN) applications. These app provide access to a partner's geographic location and other private details, serving as a strategy for reducing relationship uncertainty and enhancing monitoring capabilities. However, while LBSN app offer increased information access, they also harbor the potential to disrupt relationships. Due to the functional characteristics of these applications and variations in user personality traits, their use within romantic relationships can yield significant effects. This study aims to explore the impact of employing LBSN app as a strategy for reducing uncertainty in romantic relationships. It delves into the psychological mechanisms behind the use of location tracking surveillance and incorporates attachment styles as a moderating variable for more comprehensive investigation.
This study conducted a questionnaire survey method, and 333 valid questionnaires were collected. The results revealed that definition uncertainty positively influences mutuality uncertainty and future uncertainty. Adopting location tracking surveillance positively impacts definition uncertainty and jealousy. Behavioral norm uncertainty and mutuality uncertainty and future uncertainty positively affect intimacy. Anxious attachment only moderates the effects of location tracking surveillance on definition uncertainty and future uncertainty. This research expands IES studies and enhances understanding of URT. It also offers practical advice for LBSN app developers and users facing competition
Students' Intercultural Sensitivity in an International School in Taiwan
No full textAs the educational environment becomes increasingly diverse, local and international students must develop intercultural sensitivity to interact with people from different cultures respectfully and effectively. This quantitative study aims to investigate the students' Intercultural Sensitivity level in an International school in Taiwan. The participants in this study selected by purposive sampling method were students (n=72) from an international school in a southern metropolitan area in Taiwan. The study employed Intercultural Sensitivity Questionnaires (ISQ hereafter), developed and validated by Chen and Starosta in 2000. The ISQ questionnaire uses a five-point Likert scale to determine whether there are differences in intercultural sensitivity among students based on independent variables like gender, age, and diversity of teacher in each class. The data was analyzed using SPSS 28. The results showed statistically significant differences in the level of intercultural sensitivity among students' gender, age, and diversity of teachers in each class. The results indicated that high school students possess greater intercultural sensitivity than their primary school counterparts. Additionally, this study has demonstrated that the presence of culturally diverse teachers can substantially elevate students' intercultural sensitivity
Imbalanced Data Classification Using Gaussian Mixture Model
No full textIn the field of medicine, many datasets suffer from the issue of class imbalance. Imbalanced learning significantly affects the performance of most learning algorithms, as these algorithms typically assume or expect a relatively equal distribution of instances among classes. Consequently, when faced with imbalanced datasets, these algorithms struggle to accurately capture the feature distribution, resulting in reduced accuracy in identifying minority classes within the dataset.
If such datasets are directly fed to conventional models for training, these models can achieve high accuracy by simply classifying all inputs as the majority class. However, this approach fails to effectively predict minority class instances.
When dealing with imbalanced datasets, there are two common approaches: data preprocessing and the design of new algorithms. This study adopts the former, achieving data balance through preprocessing before using traditional classification algorithms. The preprocessing involves reducing the majority classes in the dataset to achieve balance. Similarity clustering is applied to the majority class data, where similar data points are represented by a single instance, and the rest are removed. This process aims to equalize the number of instances between majority and minority classes, enabling the model to achieve higher precision (PPV), specificity, and negative predictive value (NPV) while maintaining high sensitivity.
Aligned with clinical requirements in the field of medicine, this study aims to achieve the highest Precision when Sensitivity (Recall) is above 80%, 85%, 90%, and 95%, thereby minimizing the chances of false negatives, specifically reducing the misclassification of positive patients as negative
The Advertising Effectiveness of Different Types of Co-branding Partner on the Convenience Store Delicacies
No full textIn the past, convenience stores collaborated extensively with renowned caterers to enhance the quality of their fresh food offerings through the expertise of culinary professionals. This allowed consumers to enjoy high-quality delectable food at convenience stores at any time. However, in the past year, convenience stores have expanded their collaborations beyond just culinary experts, and partnering with social media influencers has become a new trend. Whenever these collaborations result in the development of new products, it generates significant attention and boosts sales for the convenience stores. Therefore, this study aims to investigate whether consumers' "product attitudes" and "purchase intentions" regarding co-branded fresh food are influenced by different " co-branding partner type." Additionally, when convenience stores collaborate with individuals, they often use eye-catching advertising visuals in addition to news media reports. However, these visual advertisements rely solely on non-verbal cues. Hence, this study focuses on non-verbal cues within advertisements, such as " portrait " and " typeface of names," to examine whether consumers perceive varying degrees of "perceived trustworthiness " and "curiosity" based on the " co-branding partner type," consequently affecting their "product attitudes" and "purchase intentions."
Experiment one utilized social media posts featuring co-branded fresh food products with culinary experts and social media influencers. The objective was to verify the effects of two factors: "co-branding partner type" (culinary expert vs. food influencer) and "convenience store type" (7-11 vs. Family Mart) in a two-factor between-groups design. A total of 180 valid samples were collected. In experiment two, co-branded fresh food posters were used, featuring culinary experts and social media influencers, with or without the inclusion of their images and using different typeface for their names. The goal was to examine the effects of two factors: "co-branding partner type" (culinary expert vs. food influencer) and " portrait " (presence vs. absence) as well as " typeface of names" (machine-generated vs. handwritten) in a two-factor between-groups design. A total of 420 valid samples were collected. This study, conducted as an experimental design research, explores how different combinations of co-branding partners, the presence or absence of portrait, and typeface impact consumers' attitudes and purchase intentions towards co-branded fresh food through the mediation of perceived trustworthiness and curiosity. Based on a review of the literature, the following results were found:
(1)Culinary experts as co-branding partners generate higher levels of perceived trustworthiness and curiosity compared to food influencers.\ue3(2)Perceived trustworthiness and curiosity lead to higher levels of product attitudes and purchase intentions.\ue3(3)The presence of portrait in conjunction with culinary experts leads to higher levels of perceived trustworthiness compared to food influencers, but there is no significant difference in curiosity between expert and influencer co-branding partners.\ue3(4) Machine-written typeface in conjunction with culinary experts generate higher levels of perceived trustworthiness and curiosity compared to food influencers, while handwritten typeface lead to higher perceived trust only in the case of culinary experts, with no significant difference in curiosity between expert and influencer co-branding partners.\ue3(5)The presence or absence of portrait does not impact the mediation of perceived trustworthiness and curiosity.\ue3(6) Typeface do impact the mediation of perceived trustworthiness and curiosity.
This study aims to leverage advertising design to enhance sales of co-branded fresh food products with different types of co-branding partners and to provide practical guidance for convenience stores in selecting co-branding partners and creating advertisements when launching co-branded products
Study on the network structure,optical performance,and color shift of polymer stabilized cholesteric liquid crystal displays
No full textThis study investigates the network structure, optical performance, and color shift characteristics of Polymer Stabilized Cholesteric Liquid Crystal Displays (PSCT). We conducted a detailed investigation on various parameters, such as liquid crystal host, monomer concentration, polymerization temperature, and monomer type, to understand their influence on hysteresis and color shift during PSCT driving processes. The experimental results indicate that PSCT's network structure affects its optical performance. We found that a denser network structure can enhance PSCT's stability and optical performance, and the formation of the network structure is closely related to polymerization temperature and monomer concentration, which can be controlled by adjusting these parameters. Next, we studied the hysteresis and color shift characteristics of PSCT. The results demonstrate that we can increase the density of the network structure through fabrication and material selection, and a denser structure can reduce hysteresis and color shift during PSCT driving processes. In conclusion, this study delves into the optical performance, hysteresis, and color shift characteristics of PSCT, and provides a series of optimization suggestions, offering valuable reference points for its application and development
Outfit Recommendation Using Iterative Bi-Directional Instance-Based Compatibility Prediction
No full textExisting fashion recommendation studies focus primarily on recommending individual items instead of suggesting a complete outfit. In the past, Han et al. adopted the Bi-directional Long Short-Term Memory (Bi-LSTM) mechanism to generate the embedding representation of fashion items. They used it to evaluate the compatibility score of an outfit and could also recommend a missing fashion item. This thesis proposes several novel outfit recommendation mechanisms to suggest suitable revised outfits based on the outfit inputs of users. We first propose two recommendation mechanisms, one with the first and the last fashion items fixed and the other with them varied, by extending Han et al.'s approach to recommend the whole outfit. We also propose two new metrics for outfit compatibility prediction in these recommended outfits. Besides, we then strengthen the above two mechanisms through an iterative process based on the concept that an initially good outfit input will get a better recommendation result than an initially bad one. Two corresponding iterative methods are thus proposed and can gradually produce better outfits. At last, considering the item order of an outfit input may affect the recommended results, we design two corresponding approaches using the permutation of the items in outfit inputs to improve outfit compatibility. The experimental results show that the iterative and the permutation methods can indeed produce highly compatible outfits, which are better than the outfit input. We also made a questionnaire with the results indicating the outfits recommended by the proposed approaches better align with people's aesthetics and preferences