AI Data Privacy Concerns

AI Data Privacy Concerns — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  Community cloud

  A community cloud in computing is a collaborative effort in which infrastructure is shared between several organizations from a specific community with common concerns (security, compliance, jurisdiction, etc.), whether managed internally or by a third party and hosted internally or externally. This is controlled and used by a group of organizations that have shared interests. The costs are spread over fewer users than a public cloud (but more than a private cloud), so only some of the cost savings potential of cloud computing are realized. The community cloud is provisioned for use by a group of consumers from different organizations who share the same concerns (e.g., application, security, policy, and efficiency demands).

  Read more →
• 

  AI Text-to-image Tools Reviews: What Actually Works in 2026

  In search of the best AI text-to-image tool? An AI text-to-image tool is software that uses machine learning to help you get more done — it turns a rough idea into a polished result in seconds. When choosing one, weigh output quality, pricing, export formats, and how well it fits the tools you already use. Whether you are a beginner or a pro, the right AI text-to-image tool slots into your workflow and pays for itself fast. Below we compare features, pricing, and real output so you can choose with confidence.

  Read more →
• 

  Sophia Ananiadou

  Sophia Ananiadou is a Greek-British computer scientist and computational linguist. She led the development of and directs the National Centre for Text Mining (NaCTeM) in the United Kingdom. She is also Professor in Computer Science in the Department of Computer Science at the University of Manchester. Her research focusses on biomedical text mining and natural language processing and has fed into the development of numerous applications that, for example, facilitate the discovery of new knowledge, enable exploration of historical archives, allow semantic search of biomedical literature, reduce human effort in screening search hits for production of systematic reviews, enable enrichment of metabolic pathway models with evidence from the literature, allow discovery of risk in the construction industry from health and safety incident reports and enable interoperability of components in text mining workflows. == Education == Ananiadou was educated at the Lycée français St Joseph in Athens, Greece (1969–1975). She received a Bachelor of Arts (Ptychion) from the University of Athens (1979), a Master of Advanced Studies (DEA) in Linguistics from Paris VII, Jussieu, France (1980), a DEA in Literature from Paris IV, Sorbonne, France (1984) and a PhD in Computational linguistics from the University of Manchester Institute of Science and Technology (UMIST), in 1988. == Career and research == Ananiadou was a research assistant at Dalle Molle Institute for Semantic and Cognitive Studies (ISSCO, 1983–1984), a research assistant (1985–1988) then research associate (1988–1993) in the department of language engineering at UMIST, senior lecturer at Manchester Metropolitan University (1993–1999), senior lecturer then reader in the School of Computing Science and Engineering, University of Salford (2000–2005), then reader in the School of Computer Science, University of Manchester (2005–2009). Since 2009, she has served as professor in computer science in the Department of Computer Science at the University of Manchester. In July 2025, she became deputy director of the Christabel Pankhurst Institute for health technology research and innovation, University of Manchester. From 2018–2026, she served as the deputy director of the Institute for Data Science and Artificial Intelligence, University of Manchester. She is a senior lead researcher of the ARCHIMEDES research unit of the Athena Research Centre, Greece. ARCHIMEDES is a research and innovation hub fostering international collaboration and knowledge exchange on Artificial Intelligence and Data Science. On February 7, 2025, she was appointed a member of the Artificial Intelligence Sectoral Scientific Council of the Greek Ministry of Development (announcement of appointment in Greek). She is also a Visiting Distinguished Research Fellow in the Knowledge and Information Research Team at the Artificial Intelligence Research Center (AIRC), Japan, which is a research unit of the Japanese National Institute of Advanced Industrial Science and Technology (AIST). In addition, she was appointed to the honorary position of Adjunct Professor of Wuhan University, People's Republic of China, for the period October 2025 to October 2028, collaborating with the School of Artificial Intelligence. Ananiadou has published since 1986, has an h-index of 81 and a Research.com United Kingdom ranking in Computer Science of 104. She is also ranked number 1 internationally in text mining by ScholarGPS. In addition, she is included in the Stanford/Elsevier Top 2% Scientist Rankings for 2025. Ananiadou received a Diplôme de traducteur (Diploma of Translator) from the Institut français d'Athènes, Greece (1979) and a Certificate in Counselling from the University of Salford, UK (2004). === Awards and honours === In 2019, in recognition of her contributions in Artificial Intelligence and text mining for Biomedicine, Ananiadou received an honorary doctorate from the University of the Aegean, on the 20th anniversary of its Department of Mediterranean Studies, Rhodes. Ananiadou received the Unstructured Information Management Architecture (UIMA) innovation award from IBM three years running (2006, 2007 & 2008). She was awarded the Daiwa Adrian Prize in 2004 and also received a Japan Trust award from the Ministry of Education, Japan in 1997. Ananiadou was a Turing Fellow of the Alan Turing Institute in London from 2018 to 2023. Since 2021, she is a member and, since 2024, a Fellow, of the ELLIS Society, the professional society of the cross-national European Laboratory for Learning and Intelligent Systems. Ananiadou served as vice president (VP) of the European Association for Terminology from 1997 to 1999. At the 28th International Conference on Computational Linguistics (COLING 2020), she received, with M. Li and H. Takamura, an Outstanding Paper designation for the paper "A Neural Model for Aggregating Coreference Annotation in Crowdsourcing".

  Read more →
• 

  Regularization perspectives on support vector machines

  Within mathematical analysis, Regularization perspectives on support-vector machines provide a way of interpreting support-vector machines (SVMs) in the context of other regularization-based machine-learning algorithms. SVM algorithms categorize binary data, with the goal of fitting the training set data in a way that minimizes the average of the hinge-loss function and L2 norm of the learned weights. This strategy avoids overfitting via Tikhonov regularization and in the L2 norm sense and also corresponds to minimizing the bias and variance of our estimator of the weights. Estimators with lower Mean squared error predict better or generalize better when given unseen data. Specifically, Tikhonov regularization algorithms produce a decision boundary that minimizes the average training-set error and constrain the Decision boundary not to be excessively complicated or overfit the training data via a L2 norm of the weights term. The training and test-set errors can be measured without bias and in a fair way using accuracy, precision, Auc-Roc, precision-recall, and other metrics. Regularization perspectives on support-vector machines interpret SVM as a special case of Tikhonov regularization, specifically Tikhonov regularization with the hinge loss for a loss function. This provides a theoretical framework with which to analyze SVM algorithms and compare them to other algorithms with the same goals: to generalize without overfitting. SVM was first proposed in 1995 by Corinna Cortes and Vladimir Vapnik, and framed geometrically as a method for finding hyperplanes that can separate multidimensional data into two categories. This traditional geometric interpretation of SVMs provides useful intuition about how SVMs work, but is difficult to relate to other machine-learning techniques for avoiding overfitting, like regularization, early stopping, sparsity and Bayesian inference. However, once it was discovered that SVM is also a special case of Tikhonov regularization, regularization perspectives on SVM provided the theory necessary to fit SVM within a broader class of algorithms. This has enabled detailed comparisons between SVM and other forms of Tikhonov regularization, and theoretical grounding for why it is beneficial to use SVM's loss function, the hinge loss. == Theoretical background == In the statistical learning theory framework, an algorithm is a strategy for choosing a function f : X → Y {\displaystyle f\colon \mathbf {X} \to \mathbf {Y} } given a training set S = { ( x 1 , y 1 ) , … , ( x n , y n ) } {\displaystyle S=\{(x_{1},y_{1}),\ldots ,(x_{n},y_{n})\}} of inputs x i {\displaystyle x_{i}} and their labels y i {\displaystyle y_{i}} (the labels are usually ± 1 {\displaystyle \pm 1} ). Regularization strategies avoid overfitting by choosing a function that fits the data, but is not too complex. Specifically: f = argmin f ∈ H { 1 n ∑ i = 1 n V ( y i , f ( x i ) ) + λ ‖ f ‖ H 2 } , {\displaystyle f={\underset {f\in {\mathcal {H}}}{\operatorname {argmin} }}\left\{{\frac {1}{n}}\sum _{i=1}^{n}V(y_{i},f(x_{i}))+\lambda \|f\|_{\mathcal {H}}^{2}\right\},} where H {\displaystyle {\mathcal {H}}} is a hypothesis space of functions, V : Y × Y → R {\displaystyle V\colon \mathbf {Y} \times \mathbf {Y} \to \mathbb {R} } is the loss function, ‖ ⋅ ‖ H {\displaystyle \|\cdot \|_{\mathcal {H}}} is a norm on the hypothesis space of functions, and λ ∈ R {\displaystyle \lambda \in \mathbb {R} } is the regularization parameter. When H {\displaystyle {\mathcal {H}}} is a reproducing kernel Hilbert space, there exists a kernel function K : X × X → R {\displaystyle K\colon \mathbf {X} \times \mathbf {X} \to \mathbb {R} } that can be written as an n × n {\displaystyle n\times n} symmetric positive-definite matrix K {\displaystyle \mathbf {K} } . By the representer theorem, f ( x i ) = ∑ j = 1 n c j K i j , and ‖ f ‖ H 2 = ⟨ f , f ⟩ H = ∑ i = 1 n ∑ j = 1 n c i c j K ( x i , x j ) = c T K c . {\displaystyle f(x_{i})=\sum _{j=1}^{n}c_{j}\mathbf {K} _{ij},{\text{ and }}\|f\|_{\mathcal {H}}^{2}=\langle f,f\rangle _{\mathcal {H}}=\sum _{i=1}^{n}\sum _{j=1}^{n}c_{i}c_{j}K(x_{i},x_{j})=c^{T}\mathbf {K} c.} == Special properties of the hinge loss == The simplest and most intuitive loss function for categorization is the misclassification loss, or 0–1 loss, which is 0 if f ( x i ) = y i {\displaystyle f(x_{i})=y_{i}} and 1 if f ( x i ) ≠ y i {\displaystyle f(x_{i})\neq y_{i}} , i.e. the Heaviside step function on − y i f ( x i ) {\displaystyle -y_{i}f(x_{i})} . However, this loss function is not convex, which makes the regularization problem very difficult to minimize computationally. Therefore, we look for convex substitutes for the 0–1 loss. The hinge loss, V ( y i , f ( x i ) ) = ( 1 − y f ( x ) ) + {\displaystyle V{\big (}y_{i},f(x_{i}){\big )}={\big (}1-yf(x){\big )}_{+}} , where ( s ) + = max ( s , 0 ) {\displaystyle (s)_{+}=\max(s,0)} , provides such a convex relaxation. In fact, the hinge loss is the tightest convex upper bound to the 0–1 misclassification loss function, and with infinite data returns the Bayes-optimal solution: f b ( x ) = { 1 , p ( 1 ∣ x ) > p ( − 1 ∣ x ) , − 1 , p ( 1 ∣ x ) < p ( − 1 ∣ x ) . {\displaystyle f_{b}(x)={\begin{cases}1,&p(1\mid x)>p(-1\mid x),\\-1,&p(1\mid x) Read more →

• 

  Separable filter

  A separable filter in image processing can be written as product of two more simple filters. Typically a 2-dimensional convolution operation is separated into two 1-dimensional filters. This reduces the computational costs on an N × M {\displaystyle N\times M} image with a m × n {\displaystyle m\times n} filter from O ( M ⋅ N ⋅ m ⋅ n ) {\displaystyle {\mathcal {O}}(M\cdot N\cdot m\cdot n)} down to O ( M ⋅ N ⋅ ( m + n ) ) {\displaystyle {\mathcal {O}}(M\cdot N\cdot (m+n))} . == Examples == 1. A two-dimensional smoothing filter: 1 3 [ 1 1 1 ] ∗ 1 3 [ 1 1 1 ] = 1 9 [ 1 1 1 1 1 1 1 1 1 ] {\displaystyle {\frac {1}{3}}{\begin{bmatrix}1\\1\\1\end{bmatrix}}{\frac {1}{3}}{\begin{bmatrix}1&1&1\end{bmatrix}}={\frac {1}{9}}{\begin{bmatrix}1&1&1\\1&1&1\\1&1&1\end{bmatrix}}} 2. Another two-dimensional smoothing filter with stronger weight in the middle: 1 4 [ 1 2 1 ] ∗ 1 4 [ 1 2 1 ] = 1 16 [ 1 2 1 2 4 2 1 2 1 ] {\displaystyle {\frac {1}{4}}{\begin{bmatrix}1\\2\\1\end{bmatrix}}{\frac {1}{4}}{\begin{bmatrix}1&2&1\end{bmatrix}}={\frac {1}{16}}{\begin{bmatrix}1&2&1\\2&4&2\\1&2&1\end{bmatrix}}} 3. The Sobel operator, used commonly for edge detection: [ 1 2 1 ] ∗ [ 1 0 − 1 ] = [ 1 0 − 1 2 0 − 2 1 0 − 1 ] {\displaystyle {\begin{bmatrix}1\\2\\1\end{bmatrix}}{\begin{bmatrix}1&0&-1\end{bmatrix}}={\begin{bmatrix}1&0&-1\\2&0&-2\\1&0&-1\end{bmatrix}}} This works also for the Prewitt operator. In the examples, there is a cost of 3 multiply–accumulate operations for each vector which gives six total (horizontal and vertical). This is compared to the nine operations for the full 3x3 matrix. Another notable example of a separable filter is the Gaussian blur whose performance can be greatly improved the bigger the convolution window becomes.

  Read more →
• 

  GLIMMER

  In bioinformatics, GLIMMER (Gene Locator and Interpolated Markov ModelER) is used to find genes in prokaryotic DNA. "It is effective at finding genes in bacteria, archea, viruses, typically finding 98-99% of all relatively long protein coding genes". GLIMMER was the first system that used the interpolated Markov model to identify coding regions. The GLIMMER software is open source and is maintained by Steven Salzberg, Art Delcher, and their colleagues at the Center for Computational Biology at Johns Hopkins University. The original GLIMMER algorithms and software were designed by Art Delcher, Simon Kasif and Steven Salzberg and applied to bacterial genome annotation in collaboration with Owen White. == Versions == === GLIMMER 1.0 === First Version of GLIMMER "i.e., GLIMMER 1.0" was released in 1998 and it was published in the paper Microbial gene identification using interpolated Markov model. Markov models were used to identify microbial genes in GLIMMER 1.0. GLIMMER considers the local composition sequence dependencies which makes GLIMMER more flexible and more powerful when compared to fixed-order Markov model. There was a comparison made between interpolated Markov model used by GLIMMER and fifth order Markov model in the paper Microbial gene identification using interpolated Markov models. "GLIMMER algorithm found 1680 genes out of 1717 annotated genes in Haemophilus influenzae where fifth order Markov model found 1574 genes. GLIMMER found 209 additional genes which were not included in 1717 annotated genes where fifth order Markov model found 104 genes."' === GLIMMER 2.0 === Second Version of GLIMMER i.e., GLIMMER 2.0 was released in 1999 and it was published in the paper Improved microbial identification with GLIMMER. This paper provides significant technical improvements such as using interpolated context model instead of interpolated Markov model and resolving overlapping genes which improves the accuracy of GLIMMER. Interpolated context models are used instead of interpolated Markov model which gives the flexibility to select any base. In interpolated Markov model probability distribution of a base is determined from the immediate preceding bases. If the immediate preceding base is irrelevant amino acid translation, interpolated Markov model still considers the preceding base to determine the probability of given base where as interpolated context model which was used in GLIMMER 2.0 can ignore irrelevant bases. False positive predictions were increased in GLIMMER 2.0 to reduce the number of false negative predictions. Overlapped genes are also resolved in GLIMMER 2.0. Various comparisons between GLIMMER 1.0 and GLIMMER 2.0 were made in the paper Improved microbial identification with GLIMMER which shows improvement in the later version. "Sensitivity of GLIMMER 1.0 ranges from 98.4 to 99.7% with an average of 99.1% where as GLIMMER 2.0 has a sensitivity range from 98.6 to 99.8% with an average of 99.3%. GLIMMER 2.0 is very effective in finding genes of high density. The parasite Trypanosoma brucei, responsible for causing African sleeping sickness is being identified by GLIMMER 2.0" === GLIMMER 3.0 === Third version of GLIMMER, "GLIMMER 3.0" was released in 2007 and it was published in the paper Identifying bacterial genes and endosymbiont DNA with Glimmer. This paper describes several major changes made to the GLIMMER system including improved methods to identify coding regions and start codon. Scoring of ORF in GLIMMER 3.0 is done in reverse order i.e., starting from stop codon and moves back towards the start codon. Reverse scanning helps in identifying the coding portion of the gene more accurately which is contained in the context window of IMM. GLIMMER 3.0 also improves the generated training set data by comparing the long-ORF with universal amino acid distribution of widely disparate bacterial genomes."GLIMMER 3.0 has an average long-ORF output of 57% for various organisms where as GLIMMER 2.0 has an average long-ORF output of 39%." GLIMMER 3.0 reduces the rate of false positive predictions which were increased in GLIMMER 2.0 to reduce the number of false negative predictions. "GLIMMER 3.0 has a start-site prediction accuracy of 99.5% for 3'5' matches where as GLIMMER 2.0 has 99.1% for 3'5' matches. GLIMMER 3.0 uses a new algorithm for scanning coding regions, a new start site detection module, and architecture which integrates all gene predictions across an entire genome." Minimum description length === Theoretical and Biological Foundation === The GLIMMER project helped introduce and popularize the use of variable length models in Computational Biology and Bioinformatics that subsequently have been applied to numerous problems such as protein classification and others. Variable length modeling was originally pioneered by information theorists and subsequently ingeniously applied and popularized in data compression (e.g. Ziv-Lempel compression). Prediction and compression are intimately linked using Minimum Description Length Principles. The basic idea is to create a dictionary of frequent words (motifs in biological sequences). The intuition is that the frequently occurring motifs are likely to be most predictive and informative. In GLIMMER the interpolated model is a mixture model of the probabilities of these relatively common motifs. Similarly to the development of HMMs in Computational Biology, the authors of GLIMMER were conceptually influenced by the previous application of another variant of interpolated Markov models to speech recognition by researchers such as Fred Jelinek (IBM) and Eric Ristad (Princeton). The learning algorithm in GLIMMER is different from these earlier approaches. == Access == GLIMMER can be downloaded from The Glimmer home page (requires a C++ compiler). Alternatively, an online version is hosted by NCBI [1]. == How it works == GLIMMER primarily searches for long-ORFS. An open reading frame might overlap with any other open reading frame which will be resolved using the technique described in the sub section. Using these long-ORFS and following certain amino acid distribution GLIMMER generates training set data. Using these training data, GLIMMER trains all the six Markov models of coding DNA from zero to eight order and also train the model for noncoding DNA GLIMMER tries to calculate the probabilities from the data. Based on the number of observations, GLIMMER determines whether to use fixed order Markov model or interpolated Markov model. If the number of observations are greater than 400, GLIMMER uses fixed order Markov model to obtain there probabilities. If the number of observations are less than 400, GLIMMER uses interpolated Markov model which is briefly explained in the next sub section. GLIMMER obtains score for every long-ORF generated using all the six coding DNA models and also using non-coding DNA model. If the score obtained in the previous step is greater than a certain threshold then GLIMMER predicts it to be a gene. The steps explained above describes the basic functionality of GLIMMER. There are various improvements made to GLIMMER and some of them are described in the following sub-sections. === The GLIMMER system === GLIMMER system consists of two programs. First program called build-imm, which takes an input set of sequences and outputs the interpolated Markov model as follows. The probability for each base i.e., A,C,G,T for all k-mers for 0 ≤ k ≤ 8 is computed. Then, for each k-mer, GLIMMER computes weight. New sequence probability is computed as follows. where n is the length of the sequence S x {\displaystyle S_{x}} is the oligomer at position x. I M M 8 ( S x ) {\displaystyle IMM_{8}(S_{x})} , the 8 t h {\displaystyle 8^{th}} -order interpolated Markov model score is computed as "where Y k ( S x − 1 ) {\displaystyle Y_{k}(S_{x-1})} is the weight of the k-mer at position x-1 in the sequence S and P k ( S x ) {\displaystyle P_{k}(S_{x})} is the estimate obtained from the training data of the probability of the base located at position x in the k t h {\displaystyle k^{th}} -order model." The probability of base S x {\displaystyle S_{x}} given the i previous bases is computed as follows. "The value of Y i ( S x ) {\displaystyle Y_{i}(S_{x})} associated with P i ( S x ) {\displaystyle P_{i}(S_{x})} can be regarded as a measure of confidence in the accuracy of this value as an estimate of the true probability. GLIMMER uses two criteria to determine Y i ( S x ) {\displaystyle Y_{i}(S_{x})} . The first of these is simple frequency occurrence in which the number of occurrences of context string S x , i {\displaystyle S_{x,i}} in the training data exceeds a specific threshold value, then Y i ( S x ) {\displaystyle Y_{i}(S_{x})} is set to 1.0. The current default value for threshold is 400, which gives 95% confidence. When there are insufficient sample occurrences of a context string, build-imm employ additional criteria to determine Y {\displaystyle Y} value. For a

  Read more →
• 

  Tree transducer

  In theoretical computer science and formal language theory, a tree transducer (TT) is an abstract machine taking as input a tree, and generating output – generally other trees, but models producing words or other structures exist. Roughly speaking, tree transducers extend tree automata in the same way that word transducers extend word automata. Manipulating tree structures instead of words enable TT to model syntax-directed transformations of formal or natural languages. However, TT are not as well-behaved as their word counterparts in terms of algorithmic complexity, closure properties, etcetera. In particular, most of the main classes are not closed under composition. The main classes of tree transducers are: == Top-Down Tree Transducers (TOP) == A TOP T is a tuple (Q, Σ, Γ, I, δ) such that: Q is a finite set, the set of states; Σ is a finite ranked alphabet, called the input alphabet; Γ is a finite ranked alphabet, called the output alphabet; I is a subset of Q, the set of initial states; and δ is a set of rules of the form q ( f ( x 1 , … , x n ) ) → u {\displaystyle q(f(x_{1},\dots ,x_{n}))\to u} , where f is a symbol of Σ, n is the arity of f, q is a state, and u is a tree on Γ and Q × 1.. n {\displaystyle Q\times 1..n} , such pairs being nullary. === Examples of rules and intuitions on semantics === For instance, q ( f ( x 1 , … , x 3 ) ) → g ( a , q ′ ( x 1 ) , h ( q ″ ( x 3 ) ) ) {\displaystyle q(f(x_{1},\dots ,x_{3}))\to g(a,q'(x_{1}),h(q''(x_{3})))} is a rule – one customarily writes q ( x i ) {\displaystyle q(x_{i})} instead of the pair ( q , x i ) {\displaystyle (q,x_{i})} – and its intuitive semantics is that, under the action of q, a tree with f at the root and three children is transformed into g ( a , q ′ ( x 1 ) , h ( q ″ ( x 3 ) ) ) {\displaystyle g(a,q'(x_{1}),h(q''(x_{3})))} where, recursively, q ′ ( x 1 ) {\displaystyle q'(x_{1})} and q ″ ( x 3 ) {\displaystyle q''(x_{3})} are replaced, respectively, with the application of q ′ {\displaystyle q'} on the first child and with the application of q ″ {\displaystyle q''} on the third. === Semantics as term rewriting === The semantics of each state of the transducer T, and of T itself, is a binary relation between input trees (on Σ) and output trees (on Γ). A way of defining the semantics formally is to see δ {\displaystyle \delta } as a term rewriting system, provided that in the right-hand sides the calls are written in the form q ( x i ) {\displaystyle q(x_{i})} , where states q are unary symbols. Then the semantics [ [ q ] ] {\displaystyle [\![q]\!]} of a state q is given by [ [ q ] ] = { u ↦ v ∣ u is a tree on Σ , v is a tree on Γ , and q ( u ) → δ ∗ v } . {\displaystyle [\![q]\!]=\{u\mapsto v\mid u{\text{ is a tree on }}\Sigma ,\ v{\text{ is a tree on }}\Gamma {\text{, and }}q(u)\to _{\delta }^{}v\}.} The semantics of T is then defined as the union of the semantics of its initial states: [ [ T ] ] = ⋃ q ∈ I [ [ q ] ] . {\displaystyle [\![T]\!]=\bigcup _{q\in I}[\![q]\!].} === Determinism and domain === As with tree automata, a TOP is said to be deterministic (abbreviated DTOP) if no two rules of δ share the same left-hand side, and there is at most one initial state. In that case, the semantics of the DTOP is a partial function from input trees (on Σ) to output trees (on Γ), as are the semantics of each of the DTOP's states. The domain of a transducer is the domain of its semantics. Likewise, the image of a transducer is the image of its semantics. === Properties of DTOP === DTOP are not closed under union: this is already the case for deterministic word transducers. The domain of a DTOP is a regular tree language. Furthermore, the domain is recognisable by a deterministic top-down tree automaton (DTTA) of size at most exponential in that of the initial DTOP. That the domain is DTTA-recognizable is not surprising, considering that the left-hand sides of DTOP rules are the same as for DTTA. As for the reason for the exponential explosion in the worst case (that does not exist in the word case), consider the rule q ( f ( x 1 , x 2 ) ) → g ( p 1 ( x 1 ) , p 2 ( x 1 ) , p 3 ( x 2 ) ) {\displaystyle q(f(x_{1},x_{2}))\to g(p_{1}(x_{1}),p_{2}(x_{1}),p_{3}(x_{2}))} . In order for the computation to succeed, it must succeed for both children. That means that the right child must be in the domain of p 3 {\displaystyle p_{3}} . As for the left child, it must be in the domain of both p 1 {\displaystyle p_{1}} and p 2 {\displaystyle p_{2}} . Generally, since subtrees can be copied, a single subtree can be evaluated by multiple states during a run, despite the determinism, and unlike DTTA. Thus the construction of the DTTA recognising the domain of a DTOP must account for sets of states and compute the intersections of their domains, hence the exponential. In the special case of linear DTOP, that is to say DTOP where each x i {\displaystyle x_{i}} appears at most once in the right-hand side of each rule, the construction is linear in time and space. The image of a DTOP is not a regular tree language. Consider the transducer coding the transformation f ( x ) → g ( x , x ) {\displaystyle f(x)\to g(x,x)} ; that is, duplicate the child of the input. This is easily done by a rule q ( f ( x 1 ) ) → g ( p ( x 1 ) , p ( x 1 ) ) {\displaystyle q(f(x_{1}))\to g(p(x_{1}),p(x_{1}))} , where p encodes the identity. Then, absent any restrictions on the first child of the input, the image is a classical non-regular tree language. However, the domain of a DTOP cannot be restricted to a regular tree language. That is to say, given a DTOP T and a language L, one cannot in general build a DTOP T ′ {\displaystyle T'} such that the semantics of T ′ {\displaystyle T'} is that of T, restricted to L. This property is linked to the reason deterministic top-down tree automata are less expressive than bottom-up automata: once you go down a given path, information from other paths is inaccessible. Consider the transducer coding the transformation f ( x , y ) → y {\displaystyle f(x,y)\to y} ; that is, output the right child of the input. This is easily done by a rule q ( f ( x 1 , x 2 ) ) → p ( x 2 ) {\displaystyle q(f(x_{1},x_{2}))\to p(x_{2})} , where p encodes the identity. Now let's say we want to restrict this transducer to the finite (and thus, in particular, regular) domain { f ( c , a ) , f ( c , b ) } {\displaystyle \{f(c,a),\ f(c,b)\}} . We must use the rules q ( f ( x 1 , x 2 ) ) → p ( x 2 ) , p ( a ) → a , p ( b ) → b {\displaystyle q(f(x_{1},x_{2}))\to p(x_{2}),\ p(a)\to a,\ p(b)\to b} . But in the first rule, x 1 {\displaystyle x_{1}} does not appear at all, since nothing is produced from the left child. Thus, it is not possible to test that the left child is c. In contrast, since we produce from the right child, we can test that it is a or b. In general, the criterion is that DTOP cannot test properties of subtrees from which they do not produce output. DTOP are not closed under composition. However this problem can be solved by the addition of a lookahead: a tree automaton, coupled to the transducer, that can perform tests on the domain which the transducer is incapable of. This follows from the point about domain restriction: composing the DTOP encoding identity on { f ( c , a ) , f ( c , b ) } {\displaystyle \{f(c,a),\ f(c,b)\}} with the one encoding f ( x , y ) → y {\displaystyle f(x,y)\to y} must yield a transducer with the semantics { f ( c , a ) ↦ a , f ( c , b ) ↦ b } {\displaystyle \{f(c,a)\mapsto a,\ f(c,b)\mapsto b\}} , which we know is not expressible by a DTOP. The typechecking problem—testing whether the image of a regular tree language is included in another regular tree language—is decidable. The equivalence problem—testing whether two DTOP define the same functions—is decidable. == Bottom-Up Tree Transducers (BOT) == As in the simpler case of tree automata, bottom-up tree transducers are defined similarly to their top-down counterparts, but proceed from the leaves of the tree to the root, instead of from the root to the leaves. Thus the main difference is in the form of the rules, which are of the form f ( q 1 ( x 1 ) , … , q n ( x n ) ) → q ( u ) {\displaystyle f(q_{1}(x_{1}),\dots ,q_{n}(x_{n}))\to q(u)} .

  Read more →
• 

  Jiliang Tang

  Jiliang Tang is a Chinese-born computer scientist and a University Foundation Professor of Computer Science and Engineering at Michigan State University, where he is the director of the Data Science and Engineering (DSE) Lab. His research expertise is in data mining and machine learning. == Education and career == He received his BEng in software engineering (2008) and MSc in computer science (2010) from the Beijing Institute of Technology, Beijing, China. His PhD is from Arizona State University (2015), under the direction of Huan Liu. After gaining his PhD, he worked as a research scientist at Yahoo Labs (2015–16) before joining Michigan State University as an assistant professor (2016). His research has mostly been published jointly with Huan Liu. It has received over thirteen thousand citations documented by Google Scholar, and has received coverage in the media. == Awards == He has received the 2020 ACM SIGKDD Rising Star Award that "aims to celebrate the early accomplishments of the SIGKDD communities' brightest new minds", NSF Career Award, and Michigan State University's Distinguished Withrow Research Award. == Selected publications == === Books === Jiliang Tang, Huan Liu. Trust in Social Media, (Synthesis digital library of engineering and computer science; Synthesis lectures on information security, privacy, and trust, # 13) Morgan & Claypool, 2015 ISBN 9781627054058 === Peer reviewed journal articles === Shu K, Sliva A, Wang S, Tang J, Liu H. Fake news detection on social media: A data mining perspective. ACM SIGKDD explorations newsletter. 2017 Sep 1;19(1):22-36. [1] Tang J, Alelyani S, Liu H. Feature selection for classification: A review. Data classification: Algorithms and applications. 2014:37. [2] Li J, Cheng K, Wang S, Morstatter F, Trevino RP, Tang J, Liu H. Feature selection: A data perspective. ACM Computing Surveys (CSUR). 2017 Dec 6;50(6):1-45. [3] Chang S, Han W, Tang J, Qi GJ, Aggarwal CC, Huang TS. Heterogeneous network embedding via deep architectures. InProceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining 2015 Aug 10 (pp. 119–128) Gao H, Tang J, Hu X, Liu H. Exploring temporal effects for location recommendation on location-based social networks. InProceedings of the 7th ACM conference on Recommender systems 2013 Oct 12 (pp. 93–100). Hu X, Tang J, Gao H, Liu H. Unsupervised sentiment analysis with emotional signals. InProceedings of the 22nd international conference on World Wide Web 2013 May 13 (pp. 607–618).

  Read more →
• 

  List of data science software

  This is a list of data science software and platforms used in data science, which includes programming languages, programming environments, machine learning frameworks, data engineering tools, statistical software, data analysis, plotting, MLOps systems, and more. == Programming languages == == Development environments == These interactive notebooks, IDEs, and platforms provide specialised development environments. Apache Zeppelin Architect — Eclipse (software) CoCalc Dataiku Data Science Studio FreeMat GNU Octave Google Colab DataSpell Jupyter Notebook / JupyterLab Kaggle Notebooks MATLAB O-Matrix PyCharm RStudio SAS (software) and SAS Studio Spyder Visual Studio Code == Machine and deep learning software == The Machine learning / deep learning tools support development in those fields. == Data engineering == Examples of Data engineering tools. Apache Airflow Apache Flink Apache Hadoop Apache Kafka Apache NiFi Apache Spark Dask Data build tool (dbt) == Data mining == Examples of Data mining tools. === Free and open-source === === Proprietary === == Database management == === List of RDBMS === ==== Proprietary ==== == Data warehouses == Data warehouse environments include: Amazon Redshift Snowflake Google BigQuery Microsoft Azure Synapse Teradata Vertica == Data lakes == Data lake environments include: Apache Hadoop Cloudera Databricks Delta Lake Amazon S3 Google Cloud Storage Azure Data Lake == Algorithms == Apriori algorithm – frequent itemset mining and association rule learning in market basket analysis Backpropagation – algorithm for training artificial neural networks using gradient descent Decision Trees – tree-based algorithm for classification and regression Expectation–maximization algorithm – iterative procedure for maximum likelihood estimation with latent variables Gradient descent – iterative optimization algorithm for minimizing a loss function ID3 algorithm – used to generate a decision tree from a dataset K-Means – clustering algorithm based on minimizing within-cluster distances K-Nearest Neighbors (KNN) – instance-based learning and classification method Linear regression – estimation method for predicting a dependent variable based on independent variables Logistic regression – classification algorithm for predicting a binary outcome Naive Bayes – probabilistic classifier based on Bayes' theorem Ordinary least squares – estimation method for parameters in linear regression PageRank – graph-based algorithm for link analysis and search ranking Principal component analysis – technique to reduce high-dimensional data while preserving variance Q-learning – reinforcement learning algorithm for learning optimal actions Random forest – ensemble of decision trees for improved classification or regression Sequential minimal optimization – solver for training support vector machines Stochastic gradient descent – randomized variant of gradient descent for large-scale machine learning Support Vector Machines (SVM) – algorithm for finding a hyperplane to separate classes == Statistical software == === Open-source === === Public domain === CSPro Dataplot Epi Map X-13ARIMA-SEATS === Freeware === BV4.1 MINUIT WinBUGS Winpepi === Proprietary === == Data processing == Tools for Data processing and analysis: == Data and information visualization == Software for Data visualization: == Plotting software == Software for plotting data to support processing and visualise results. == Maps and geospatial visualization == ArcGIS Carto Epi Map GeoDA Google Earth Engine Leaflet Mapbox MountainsMap QGIS == Machine learning == MLOps and model deployment: BentoML Data Version Control (DVC) Kubeflow MLflow Seldon Core Streamlit TensorFlow Serving Weights & Biases == Data repositories == Kaggle – platform for data science competitions, datasets, and notebooks. OpenML – collaborative platform for sharing datasets, algorithms, and experiments. University of California, Irvine Machine Learning Repository Zenodo – open-access repository supported by CERN and the EU. == Educational data science software == Kaggle – online platform for data science education, competitions, datasets, and collaborative learning. KNIME – open-source data analytics platform used for teaching data science, machine learning, and workflow-based analysis. RapidMiner – used in academic research and education for data mining and machine learning. Statistics Online Computational Resource (SOCR) – online tools and instructional resources for statistics education. Tanagra (machine learning) – data mining software developed for research and teaching purposes. TinkerPlots – explore and analyze data through visual modeling.

  Read more →
• 

  Co-occurrence

  In linguistics, co-occurrence or cooccurrence (in older texts often shown with diacritic as coöccurrence) is an above-chance frequency of ordered occurrence of two adjacent terms in a text corpus. Co-occurrence in this linguistic sense can be interpreted as an indicator of semantic proximity or an idiomatic expression. Corpus linguistics and its statistical analyses can reveal (regularity of) patterns of co-occurrences within a language and enable the working out of typical collocations for its lexical items. A co-occurrence restriction is identified when linguistic elements never occur together. Analysis of these restrictions can lead to discoveries about the structure and development of a language. Co-occurrence can be seen an extension of word counting in higher dimensions. Co-occurrence can be quantitatively described using measures like a massive correlation or mutual information. Co-occurrence information and knowledge of co-occurring words may be relevant in analysis of language for the purposes of large language models, part of the emerging field of artificial intelligence, and helpful in word games such as scrabble.

  Read more →
• 

  Multiline optical-character reader

  A multiline optical-character reader, or MLOCR, is a type of mail sorting machine that uses optical character recognition (OCR) technology to determine how to route mail through the postal system. MLOCRs work by capturing images of the front of letter-sized mailpieces, and extracting the entire address from each piece. It looks up the postal code within each address in a master database, prints a barcode representing this information on the mailpiece, and performs an initial sort. All of this occurs in a fraction of a second as the mailpiece passes through the machine. After this point, mail is further sorted by barcode sorters that read this barcode to determine its destination throughout its journey all the way down to the walk sequence of the mail carrier. The United States Postal Service has used remote bar coding since 1992. In the United States, if the MLOCR is not able to decode the address, then the mailpiece is placed on "hold" by printing a unique fluorescent barcode on the back of the mailpiece, and the mailpiece is then set aside for further processing by the Remote Bar Coding System (formerly called Remote Video Encoding). An image of the mailpiece is sent to a Remote Encoding Center where a human data conversion operator manually inspects the image. The operator converts the information on the mailpiece into abbreviated codes and enters the data into the computer. This data is sent back to the MLOCR site where it is matched with the unique barcode on the back of the un-coded mailpiece, and a barcode is then printed on the mailpiece like the rest of the mail. All this effort is invested up front into deciphering the destination of each mailpiece and printing the correct barcode, so that the mailpiece will never need to be manually examined again until it reaches the hands of the letter carrier who will carry it to the final delivery point. A Delivery Bar Code Sorter is repeatedly used at each point in the USPS system to read the barcode and sort the mailpiece to a tray corresponding to the next leg of its journey towards its final destination. The United States Postal Service is the largest user of these machines; however, large volume mailers and mail consolidators also have their own MLOCR systems to barcode outgoing mail in order to receive significant postage discounts. An option called FASTforward can be added to an MLOCR that allows it to automatically forward mail to a new address. This additional computer hardware/software combination looks up decoded addresses in the National Change of Address database to see if the recipient has recently moved. If so, a POSTNET barcode representing the new address is sprayed on the mailpiece thus routing it to new address although the old address is still visible—a testament to the degree at which mail can be mechanically sorted. Generally, all OCR-equipped letter sorting machines ordered since the late 1980s have been equipped with OCR systems capable of reading multiple lines of address.

  Read more →
• 

  François Chollet

  François Chollet (French: [fʁɑ̃swa ʃoˈlɛ]; born 20 October 1989) is a French software engineer, artificial intelligence (AI) researcher, and former Senior Staff Engineer at Google. Chollet is the creator of the Keras deep-learning library released in 2015. His research focuses on computer vision, the application of machine learning to formal reasoning, abstraction, and how to achieve greater generality in artificial intelligence (AGI). == Education and career == In 2012, Chollet graduated with a Diplôme d'Ingénieur (Master of Engineering) from ENSTA Paris, a school of the Polytechnic Institute of Paris. In 2015, Chollet started working at Google shortly after releasing Keras. In 2019, he published the Abstraction and Reasoning Corpus for Artificial General Intelligence (ARC-AGI) benchmark, which measures the ability of AI systems to solve novel reasoning problems. In 2024, Chollet launched ARC Prize, a US$1 million competition to solve the ARC-AGI benchmark. He left Google in November 2024 after more than 9 years with the company to found with Zapier co-founder Mike Knoop a new startup focused on developing AGI with program synthesis. In early 2025, Chollet announced the expansion of ARC Prize into a full-fledged non-profit foundation, to further the mission of guiding and accelerating research progress towards artificial general intelligence. == Books and publications == Chollet's research papers in artificial intelligence have been published at major conferences in the field, including the Conference on Computer Vision and Pattern Recognition (CVPR), the Conference on Neural Information Processing Systems (NeurIPS), and the International Conference on Learning Representations (ICLR). Chollet is the author of Xception: Deep Learning with Depthwise Separable Convolutions, which is among the top ten most cited papers in CVPR proceedings at more than 18,000 citations. Chollet is the author of the book Deep Learning with Python, which sold over 100,000 copies, and the co-author with Tomasz Kalinowski of Deep Learning With R. == Awards == On December 1, 2021, Chollet won the Global Swiss AI Award for breakthroughs in AI. In September 2024, Chollet was named by TIME as one of the 100 most influential people in AI.

  Read more →
• 

  VistaCreate

  VistaCreate (formerly Crello) is an online graphic design platform for non-designers, launched in 2016. As of 2022, it has more than 10 million users in 192 countries. == Overview == VistaCreate (then known as Crello) was launched in 2016 as a part of Depositphotos. In 2019, the product hit a milestone of 1 million registered users and also launched mobile apps. In 2020, the library of templates and objects became free. A music library and a background remover tool were added to the platform. In May 2021, Moufflons Basketball, in collaboration with VistaCreate, organized a poster design competition in support of gender equality in sports. In October 2021, Vistaprint acquired Crello and its parent company, Depositphotos, for a total price of $85 million. After the acquisition, Crello was rebranded to VistaCreate. Along with Vistaprint and 99designs, it became part of the new Vista parent brand. After Russia started a full-scale war on the territory of Ukraine in February 2022, VistaCreate suspended all business in Russia and Belarus. VistaCreate's team and Depositphotos gathered collections of images and templates dedicated to the war in Ukraine.

  Read more →
• 

  Is an AI Humanizer Worth It in 2026?

  Shopping for the best AI humanizer? An AI humanizer is software that uses machine learning to help you get more done — it keeps getting smarter as the underlying models improve. Pricing, accuracy, and the size of the model behind the tool are the three factors that most affect daily usefulness. Whether you are a beginner or a pro, the right AI humanizer slots into your workflow and pays for itself fast. We tested the leading options and ranked them by quality, value, and ease of use.

  Read more →
• 

  The Best Free AI Copywriting Tool for Beginners

  Curious about the best AI copywriting tool? An AI copywriting tool is software that uses machine learning to help you get more done — it combines speed, accuracy, and an interface that just works. Hands-on testing shows real-world results vary, so a short free trial is the smartest way to decide. Whether you are a beginner or a pro, the right AI copywriting tool slots into your workflow and pays for itself fast. This guide breaks down the top picks, their pros and cons, and who each one is best for.

  Read more →