In this paper we study the personalized book recommender system in a child-robot interactive environment. Firstly, we propose a novel text search algorithm using an inverse filtering mechanism that improves the efficiency. Secondly, we propose a user interest prediction method based on the Bayesian network and a novel feedback mechanism. According to children's fuzzy language input, the proposed method gives the predicted interests. Thirdly, the domain specific synonym association is proposed based on word vectorization, in order to improve the understanding of user intention. Experimental results show that the proposed recommender system has an improved performance and it can operate on embedded consumer devices with limited computational resources.

Now more than ever, we are seeing digital transformation at the most progressive institutions moving beyond mobile banking upgrades and modest changes in back office technologies to becoming organization-wide initiatives. With budgets increasing and new advanced technologies being implemented, ownership of the digital transformation process at leading banks and credit unions is moving up the organizational chart, with the more digitally mature institutions having the process led by the the CEO or even the board of directors. We are also seeing greater cross-functional engagement across market-leading institutions. Understanding how leading organizations are managing their digital transformation process is more important than ever to the future competitiveness of banks and credit unions of all sizes. Unfortunately, as we found in the area of innovation, a surprisingly small number of financial institutions (12%) consider themselves to be digital transformation'leaders', with 34% considering themselves to be'fast followers' and 55% stating they were either'mainstream players' or'laggards'.

Britain will attempt to move away from European data protection regulations as it overhauls its privacy rules after Brexit, the government has announced. The freedom to chart its own course could lead to an end to irritating cookie popups and consent requests online, said the culture secretary, Oliver Dowden, as he called for rules based on "common sense, not box-ticking". But any changes will be constrained by the need to offer a new regime that the EU deems adequate, otherwise data transfers between the UK and EU could be frozen. A new information commissioner will be put in charge of overseeing the transformation. John Edwards, currently the privacy commissioner of New Zealand, has been named as the government's preferred candidate to replace Elizabeth Denham, whose term in office will end on 31 October after a three-month extension.

Law enforcement agencies and government organizations from 24 countries outside the United States used a controversial facial recognition technology called Clearview AI, according to internal company data reviewed by BuzzFeed News. That data, which runs up until February 2020, shows that police departments, prosecutors' offices, universities, and interior ministries from around the world ran nearly 14,000 searches with Clearview AI's software. At many law enforcement agencies from Canada to Finland, officers used the software without their higher-ups' knowledge or permission. After receiving questions from BuzzFeed News, some organizations admitted that the technology had been used without leadership oversight. In March, a BuzzFeed News investigation based on Clearview AI's own internal data showed how the New York–based startup distributed its facial recognition tool, by marketing free trials for its mobile app or desktop software, to thousands of officers and employees at more than 1,800 US taxpayer-funded entities.

Offensive Language detection in social media platforms has been an active field of research over the past years. In non-native English spoken countries, social media users mostly use a code-mixed form of text in their posts/comments. This poses several challenges in the offensive content identification tasks, and considering the low resources available for Tamil, the task becomes much harder. The current study presents extensive experiments using multiple deep learning, and transfer learning models to detect offensive content on YouTube. We propose a novel and flexible approach of selective translation and transliteration techniques to reap better results from fine-tuning and ensembling multilingual transformer networks like BERT, Distil- BERT, and XLM-RoBERTa. The experimental results showed that ULMFiT is the best model for this task. The best performing models were ULMFiT and mBERTBiLSTM for this Tamil code-mix dataset instead of more popular transfer learning models such as Distil- BERT and XLM-RoBERTa and hybrid deep learning models. The proposed model ULMFiT and mBERTBiLSTM yielded good results and are promising for effective offensive speech identification in low-resourced languages.

Existing machines are functionally specific tools that were made for easy prediction and control. Tomorrow's machines may be closer to biological systems in their mutability, resilience, and autonomy. But first they must be capable of learning, and retaining, new information without repeated exposure to it. Past efforts to engineer such systems have sought to build or regulate artificial neural networks using task-specific modules with constrained circumstances of application. This has not yet enabled continual learning over long sequences of previously unseen data without corrupting existing knowledge: a problem known as catastrophic forgetting. In this paper, we introduce a system that can learn sequentially over previously unseen datasets (ImageNet, CIFAR-100) with little forgetting over time. This is accomplished by regulating the activity of weights in a convolutional neural network on the basis of inputs using top-down modulation generated by a second feed-forward neural network. We find that our method learns continually under domain transfer with sparse bursts of activity in weights that are recycled across tasks, rather than by maintaining task-specific modules. Sparse synaptic bursting is found to balance enhanced and diminished activity in a way that facilitates adaptation to new inputs without corrupting previously acquired functions. This behavior emerges during a prior meta-learning phase in which regulated synapses are selectively disinhibited, or grown, from an initial state of uniform suppression.

Several metrics have been proposed for assessing the similarity of (abstract) meaning representations (AMRs), but little is known about how they relate to human similarity ratings. Moreover, the current metrics have complementary strengths and weaknesses: some emphasize speed, while others make the alignment of graph structures explicit, at the price of a costly alignment step. In this work we propose new Weisfeiler-Leman AMR similarity metrics that unify the strengths of previous metrics, while mitigating their weaknesses. Specifically, our new metrics are able to match contextualized substructures and induce n:m alignments between their nodes. Furthermore, we introduce a Benchmark for AMR Metrics based on Overt Objectives (BAMBOO), the first benchmark to support empirical assessment of graph-based MR similarity metrics. BAMBOO maximizes the interpretability of results by defining multiple overt objectives that range from sentence similarity objectives to stress tests that probe a metric's robustness against meaning-altering and meaning-preserving graph transformations. We show the benefits of BAMBOO by profiling previous metrics and our own metrics. Results indicate that our novel metrics may serve as a strong baseline for future work.

Time series data can be subject to changes in the underlying process that generates them and, because of these changes, models built on old samples can become obsolete or perform poorly. In this work, we present a way to incorporate information about the current data distribution and its evolution across time into machine learning algorithms. Our solution is based on efficiently maintaining statistics, particularly the mean and the variance, of data features at different time resolutions. These data summarisations can be performed over the input attributes, in which case they can then be fed into the model as additional input features, or over latent representations learned by models, such as those of Recurrent Neural Networks. In classification tasks, the proposed techniques can significantly outperform the prediction capabilities of equivalent architectures with no feature / latent summarisations. Furthermore, these modifications do not introduce notable computational and memory overhead when properly adjusted.

Recommender Systems are algorithms that predict a user's preference for an item. Reciprocal Recommenders are a subset of recommender systems, where the items in question are people, and the objective is therefore to predict a bidirectional preference relation. They are used in settings such as online dating services and social networks. In particular, images provided by users are a crucial part of user preference, and one that is not exploited much in the literature. We present a novel method of interpreting user image preference history and using this to make recommendations. We train a recurrent neural network to learn a user's preferences and make predictions of reciprocal preference relations that can be used to make recommendations that satisfy both users. We show that our proposed system achieves an F1 score of 0.87 when using only photographs to produce reciprocal recommendations on a large real world online dating dataset. Our system significantly outperforms on the state of the art in both content-based and collaborative filtering systems.

As the use of interactive machines grow, the task of Emotion Recognition in Conversation (ERC) became more important. If the machine generated sentences reflect emotion, more human-like sympathetic conversations are possible. Since emotion recognition in conversation is inaccurate if the previous utterances are not taken into account, many studies reflect the dialogue context to improve the performances. We introduce CoMPM, a context embedding module (CoM) combined with a pre-trained memory module (PM) that tracks memory of the speaker's previous utterances within the context, and show that the pre-trained memory significantly improves the final accuracy of emotion recognition. We experimented on both the multi-party datasets (MELD, EmoryNLP) and the dyadic-party datasets (IEMOCAP, DailyDialog), showing that our approach achieve competitive performance on all datasets.