Music emotion recognition algorithm based on BP neural network

Music plays a very important role in animation production. Because it could better express the emotion of the character, this paper uses BP neural network to identify the music emotion. This paper first introduced the structure of BP neural network. Then, the parameters and structure of the network were designed according to the category of music emotion. Finally, a three-layer BP neural network with 5 input nodes, 13 hidden layer nodes and 4 output nodes was constructed and applied to music emotion recognition. The recognition accuracy was 85.02%, which basically met the requirements of music emotion recognition and achieves the expected effect.

Deep Reinforcement Learning Algorithms for Path Planning Domain in Grid-like Environment

Recently, more and more solutions have utilised artificial intelligence approaches in order to enhance or optimise processes to achieve greater sustainability. One of the most pressing issues is the emissions caused by cars; in this paper, the problem of optimising the route of delivery cars is tackled. In this paper, the applicability of the deep reinforcement learning algorithms with regards to the aforementioned problem is tested on a simulation game designed and implemented to pose various challenges such as constant change of delivery locations. The algorithms chosen for this task are Advantage Actor-Critic (A2C) with and without Proximal Policy Optimisation (PPO). These novel and advanced reinforcement learning algorithms have yet not been utilised in similar scenarios. The differences in performance and learning process of those are visualised and discussed. It is demonstrated that both of those algorithms present a slow but steady learning curve, which is an expected effect of reinforcement learning algorithms, leading to a conclusion that the algorithms would discover an optimal policy with an adequately long learning process. Additionally, the benefits of the Proximal Policy Optimisation algorithm are proven by the enhanced learning curve with comparison to the Advantage Actor-Critic approach, as the learning process is characterised by faster growth with a significantly smaller variation. Finally, the applicability of such algorithms in the described scenarios is discussed, alongside the possible improvements and future work.

Towards a simplified justification of the chosen sample size

Since the early work of Cohen (1962) psychological researchers have become aware of the importance of doing a power analysis to ensure that the predicted effect will be detectable with sufficient statistical power. APA guidelines require researchers to provide a justification of the chosen sample size with reference to the expected effect size; an expectation that should be based on previous research. However, we argue that a credible estimate of an expected effect size is only reasonable under two conditions: either the new study forms a direct replication of earlier work or the outcome scale makes use of meaningful and familiar units that allow for the quantification of a minimal effect of psychological interest. In practice neither of these conditions is usually met. We propose a different rationale for a power analysis that will ensure that researchers will be able to justify their sample size as meaningful and adequate.

Augment BERT with average pooling layer for Chinese summary generation

The BERT pre-trained language model has achieved good results in various subtasks of natural language processing, but its performance in generating Chinese summaries is not ideal. The most intuitive reason is that the BERT model is based on character-level composition, while the Chinese language is mostly in the form of phrases. Directly fine-tuning the BERT model cannot achieve the expected effect. This paper proposes a novel summary generation model with BERT augmented by the pooling layer. In our model, we perform an average pooling operation on token embedding to improve the model’s ability to capture phrase-level semantic information. We use LCSTS and NLPCC2017 to verify our proposed method. Experimental data shows that the average pooling model’s introduction can effectively improve the generated summary quality. Furthermore, different data needs to be set with varying pooling kernel sizes to achieve the best results through comparative analysis. In addition, our proposed method has strong generalizability. It can be applied not only to the task of generating summaries, but also to other natural language processing tasks.

Experiences and Perceptions of Default Nudges

In this thesis I argue that how people perceive and experience nudge interventions is an underappreciated factor that can help assess, and potentially address, both effectiveness and ethical concerns. In the introduction, I outline a framework for how this can be the case. Specifically, I propose that people’s perceptions and experiences are relevant to assessing (1) the ethics of nudging, (2) the net effect of nudging, and that they may (3) be moderators of the success of nudges influencing behavior. I then present three empirical studies (nine sub-studies, total N = 5171) that used online experiments to assess how people perceive and experience being subjected to nudges. In contrast to the majority of similar research, the present studies primarily subjected participants to nudges first-hand. That is, participants did not rate descriptions of nudges, but engaged in choice tasks first hand before assessing the intervention. Results indicated that people subjected to default nudges: experienced themselves as highly, and not less, autonomous and satisfied with their choices (Study I); found the use of the nudge to be fair, though less fair than no nudge (Study II); and favorably perceived a choice architect using the nudge, and not less so than in the absence of a nudge (Study III). Additionally, the studies found that making the nudge increasingly transparent by disclosing its presence and expected effect to participants did not meaningfully change their experiences of choosing, but could either improve or worsen perceptions of the nudge depending on the circumstances of the evaluation. When participants were disclosed of the nudge after they had already made a choice, this was found to negatively affect their perceptions of the choice architect. Providing a nudge disclosure did not, however, influence participants’ propensity to act in line with the nudge. I conclude that the present findings paint a generally positive picture of how default nudges are perceived and experienced, but that more research is necessary to properly inform policy. I suggest that policymakers should routinely use measures of choice experiences as a guide when designing new nudge interventions. Nevertheless, as judged by the empirical evidence available at present, default nudges appear to be a viable form of intervention in relation to concerns about both their effectiveness and ethicality.

Ameliorative effect of sodium carbonate on phytotoxicity of palm oil mill effluent

This study aims to evaluate the impact of palm oil mill effluent treated with sodium carbonate on the growth of maize. The experiment consisted of eight treatments of concentrations of palm oil mill effluent viz 1.0%, 2.0%, 3.0%, 4.0%, 5.0%, 6.0%, 7.0% and 8.0%. The treatment was carried out on potted maize plants with six replications. Four kilograms of soil was weighed into each pot. Six pots were labelled as control with untreated palm oil mill effluent added to them. Eight other groups consisting of six pots each were treated with 1.0%, 2.0%, 3.0%, 4.0%, 5.0%, 6.0%, 7.0% and 8.0% concentrations of the effluent. The maize seeds were planted one in each pot to avoid overcrowding. The pots were transferred to greenhouse and moistened daily with the effluent. The growth rate, plant height and leaf length of the maize plants were thereafter determined. The 8.0% concentration of sodium carbonate in the effluent had the highest expected effect on the parameters whereas the control gave the lowest expected effect. The effects increased as the concentration of sodium carbonate in the effluent increased, indicating that increased concentration of sodium carbonate recorded significant increase in growth rate, height and leaf length of maize. The analysis of variance for the obtained data showed that the effects of the different concentrations were significantly different. This study provides an alternative and cost effective method of ameliorating the toxicity of palm oil mill effluent to plants.

The influence of anger on empathy and theory of mind

Social cognition allows humans to understand and predict other people’s behavior by inferring or sharing their emotions, intentions and beliefs. Few studies have investigated the impact of one’s own emotional state on understanding others. Here, we tested the effect of being in an angry state on empathy and theory of mind (ToM). In a between-groups design we manipulated anger status with different paradigms in three studies (autobiographical recall (N = 45), negative feedback (N = 49), frustration (N = 46)) and checked how this manipulation affected empathic accuracy and performance in the EmpaToM. All paradigms were successful in inducing mild anger. We did not find the expected effect of anger on empathy or ToM performance but observed small behavioral changes. Together, our results validate the use of three different anger induction paradigms and speak for rather weak behavioral effects of mild state anger on empathy and ToM.

​Effect of Palm Oil Mill Effluent on the Growth of Maize (Zea mays)

Background: Raw palm oil mill effluent is toxic effluent capable of posing serious threat to plants when discharged to the environment. In developing countries, the untreated effluent is often discharged to the surrounding land due to high cost associated with its treatment. The aim of the current study was to evaluate the impact of palm oil mill effluent treated with locally available material on the growth of maize. Methods: The bunch ash obtained from local material by burning de-fruited oil palm bunch was used for the treatment. The experiment was consisted of eight treatments of concentrations of palm oil mill effluent viz 1.0%, 2.0%, 3.0%, 4.0%, 5.0%, 6.0%, 7.0% and 8.0%. The treatment was carried out on potted maize plants with six replications. Four kilograms of soil was weighed into each pot. Six pots were labelled as control, with untreated palm oil mill effluent added to them. Eight other groups consisting of six pots each were treated with 1.0%, 2.0%, 3.0%, 4.0%, 5.0%, 6.0%, 7.0% and 8.0% concentrations of the effluent. The maize seeds were planted one in each pot to avoid overcrowding. The pots were transferred to greenhouse and each pot was moistened daily with the specified concentration of the palm oil mill effluent. The growth rate, plant height and leaf length of the maize plants were thereafter determined. Result: The 8.0% concentration of bunch ash in the effluent had the highest expected effect on all the parameters whereas the control gave the lowest expected effect. The effects increased as the concentration of bunch ash in the effluent increased, indicating that increased concentration of bunch ash recorded significant increase in growth rate, height and leaf length of maize. The analysis of variance for the obtained data showed that the effects of the different concentrations were significantly different. This study provides an alternative and cost effective method of ameliorating the toxicity of palm oil mill effluent to plants.