Translation engine suggestions via target probes
The method segments text into probes and applies the best-performing translation engine to each segment, addressing the lack of granular evaluation in conventional engines, achieving improved translation accuracy by combining outputs from multiple engines.
Patent Information
- Application Number
- CN202010847482.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-09-09
- Filing Date
- 2020-08-21
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2040-08-21
AI Technical Summary
Conventional translation engines lack a granular-level evaluation method, resulting in lost or erroneous meaning in translations, and are typically evaluated based on total output rather than at the word, phrase, or sentence level.
A method that segments input text into probes (sentences, clauses, or phrases) and applies the best-performing translation engine to each segment using a Siamese RNN network and LSTM phrase reconstruction, with human expert validation and scoring to ensure semantic accuracy.
Enhances translation accuracy by identifying the best translation engine for each segment, ensuring semantically and syntactically accurate translations by combining outputs from multiple engines.
Smart Images

Figure CN112463917B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates generally to translation engine suggestion methods, and more particularly, but not by way of limitation, to systems, methods, and computer program products for using a target probe to determine a translation engine that performs better when provided with input similar to a given probe. Background Art
[0002] Often, some text needs to be translated from one language to another for communication. There are various translation engines on the Internet.
[0003] The output of conventional translation engines is often characterized by lost meaning after translation and translation resulting in erroneous meaning. Therefore, there is a lack of a baseline for comparing translations.
[0004] Furthermore, conventional translation techniques are evaluated only based on the total output of the translation rather than at a granular level (e.g., word, phrase / clause, and / or sentence level). Thus, translations are compared only based on the total output of the translation rather than on granularity.
[0005] For example, Figure 2-3 As shown in Figure 1, where "A" is the original message in English, "B" is the Swahili translation using the first translator, and "C" is the Swahili translation using the second translator, analysis of the two translations shows that in some phrases, the first translator performs better than the second translator (the first translator treats Sekou as an inanimate object, while the second translator is able to say that Sekou is a person). On the other hand, the first translator appears to distinguish between institutional names, which do not necessarily Figure 3 Translate as shown. Summary of the invention
[0006] In view of the problems in the art, the inventors have considered a new and improved technology of inserting probes into the translation engine to determine the accuracy at a granular level (for example, a probe is a sentence (or clause, phrase or word) which has a desired translation) so as to recommend a translation engine combination to translate a piece of text (for example, multiple sentences) using the best translation engine for each sentence.
[0007] In an exemplary embodiment, the present invention provides a computer-implemented translation engine suggestion method, the method comprising identifying probes of a third-party translation engine for an input text, segmenting portions of the input text into a plurality of segments based on the identified probes, dividing the input text into segments based on the segments, applying the identified probes to each segment using a corresponding third-party translation engine; and outputting a translation by combining each segment.
[0008] One or more other exemplary embodiments include a computer program product and a system based on the above-described method.
[0009] Other details and embodiments of the present invention are described below so that the present contribution to the prior art can be better understood. However, the present invention is not limited in its application to the details, wordings, terms, illustrations and / or arrangements set forth in the specification or shown in the drawings. Rather, the present invention can have embodiments in addition to the described embodiments and can be practiced and implemented in various ways and should not be considered as limiting.
[0010] Thus, those skilled in the art will appreciate that the concepts upon which this disclosure is based can be readily used as a basis for designing other structures, methods and systems to achieve the several purposes of the present invention. It is therefore important that the claims be considered to include such equivalent constructions without departing from the spirit and scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Various aspects of the present invention will be better understood through the following detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings, in which:
[0012] Figure 1 A high-level flow chart of a translation engine suggestion method 100 according to an embodiment of the present invention is exemplarily shown;
[0013] Figure 2 illustratively illustrating an example translation using a first translator engine and a second translator engine;
[0014] Figure 3 The exemplary embodiment depicts Figure 2 better translation of texts;
[0015] Figure 4 The system architecture 200 according to an embodiment of the present invention is exemplarily shown;
[0016] Figure 5 A cloud computing node 10 according to an embodiment of the present invention is shown;
[0017] Figure 6 depicts a cloud computing environment 50 according to an embodiment of the present invention; and
[0018] Figure 7 Represents an abstract model layer according to an embodiment of the present invention. DETAILED DESCRIPTION
[0019] Now refer to Figure 1-7The present invention is described, wherein the same reference numerals always represent the same parts. It is emphasized that, according to convention, the various features of the drawings are not necessarily drawn to scale. On the contrary, the size of the various features may be arbitrarily enlarged or reduced for the sake of clarity.
[0020] By Introduction Figure 1 In accordance with the illustrated example, an embodiment of the translation engine suggestion method 100 of the present invention may include various steps for determining the best translator for a given probe and recommending a combination of translators to translate a text input.
[0021] By Introduction Figure 5 In the example shown, one or more computers of the computer system 12 according to an embodiment of the present invention may include a memory 28 having instructions stored in the memory system to execute Figure 1 steps.
[0022] Although one or more embodiments may be implemented in a cloud environment 50 (e.g., Figure 7 ), but it should be understood that the present invention can be implemented outside the cloud environment.
[0023] Typically, a probe (or multiple probes) is used to evaluate the performance of a translation engine of multiple semantic types. When a new text is given to the system to be translated, it finds the probe with the highest semantic similarity for each text segment, as predicted by the Siamese RNN network. The probe is a sentence (or clause or phrase, or at the finest granularity, a word) with the desired translation.
[0024] Usually refer to Figures 1 to 4 In step 101, the current language of the input text and the user's desired translation language are identified. In step 102, relevant probes are obtained from the corpus based on the current and desired translation languages. Human experts are involved in developing and validating the probe library. A probe is a known "accurate" translation of a word, a set of words in a sentence, and / or a sentence from one language to another language. The probe is trained by running it in all available machine translation services and an output score is calculated based on the manual probe translation. An ordered list of probes and machine translation services based on the above scores are stored in the probe library.
[0025] In step 103, the input text is passed through a parser, which identifies and segments different parts based on the probes obtained. That is, step 103 breaks the input text into chunks (and the closest probes that match the chunks). The parser relies on the probes to score the similarities between the various parts of the text. The output is the text divided into several segments, each segment is annotated with a probe. In other words, the probes for the third-party translation engine used for the input text are identified by the different segments of the text (e.g., a probe and translation engine combination for each segment (multiple probe and translation engine combinations)). The present invention can access third-party translation engine application programming interfaces (APIs), including Microsoft Translator TM , Google Translate TM , WorldLingo TM , DeepL TM wait.
[0026] In step 104, the input text is segmented according to the segmentation of step 103 (e.g., different parts of the text are matched to the segments), and each fragment is applied to the best-known machine translation service with annotated probes. The "best-known machine translator with annotated probes" learns over time by using probes. For example, multiple probes are used for a segment, and then the best probe is verified. For multiple sentences, multiple translation engines can be used. And, through annotation, the present invention annotates different phrases of the text to be translated (e.g., segments) using matching probes, and applies the better-performing translation engine to each of these individually annotated phrases.
[0027] In step 105, the translated segment is passed through a long short-term memory (LSTM) phrase reconstruction engine, which outputs the translated text. That is, the best translations for each segment are combined to output the best translation across multiple translation engines using the probe (e.g., the segment is translated using multiple engines, which may perform well on different parts of the input text). In this way, the user can obtain the best translation by combining the translation API across multiple engines (i.e., determining the better performing translation engine when similar inputs are provided to a given probe).
[0028] In one embodiment, the input text of each probe is translated into an intermediate language before being translated into the final output language in step 105. Thus, the present invention learns the accuracy of the probe by including the intermediate language between translations (e.g., when translating from French to Swahili, direct translation (French to Swahili) can be avoided, and English can be added as an intermediate language (French to English to Swahili), etc.).
[0029] refer to Figure 4The present invention processes the input training text by marking it into sentences (or phrases, clauses or words), which are mapped to a vector space through sentence embedding to learn how to segment sentences into segments, and trains the model to identify the best probes. The sentences are then clustered according to the embedding. The sentences at the center of the cluster (i.e., the centroid) are identified and used as cluster labels. Human experts translate the sentences at the centroid into the target language. The centroid is then sent to different translation engines, and the translations are compared with those of human experts. The translation engines are then scored using metrics such as bilingual evaluation benches (e.g., BLEU or ROGUE). Sentence embeddings and translation metric scores are used to train a recursive neural network (RNN) semantic similarity model, which is used to map new sentences to the best translation engine.
[0030] That is, sentence embeddings are obtained by performing mean pooling on the Glove embeddings of each word of the segmented sentences from some text corpus. The embeddings are clustered using k-means clustering, where k is set to the number of probes needed to evaluate the system. The centroid of each cluster is mapped back to its original sentence, and the set of k sentences is called a probe. The probes (in the primary language) are sent to the translation engine, and for each probe and engine pair we obtain, an evaluation score (blue or out of order) is obtained by comparing with the corresponding translation (in the secondary language) by the expert. The expert also provides a subjective coherence score based on the grammatical basis of the language. The translated probes (from the engine) are then translated back to the primary language again by the expert, and these are used to augment semantic similarity datasets (such as the STS benchmark). The dataset is then used to train an RNN-based Siamese network (i.e., an artificial neural network that works together on two different input vectors simultaneously using the same weights to compute comparable output vectors) to perform semantic similarity matching settings as a binary classification task, where sentence pairs are evaluated as semantically equivalent or not. When a new sentence is given to the system, it finds the probe with the highest similarity to the one predicted by the Siamese network.
[0031] It is worth noting that the creation of probes may require human experts to generate the probe list based on their experience with the language pairs under consideration.
[0032] Another technique for obtaining probes is for human experts to generate a list of probes based on their experience with the language pair under consideration.
[0033] Experts can also set clustering criteria based on a list of probe types / markers.
[0034] like Figure 4As shown at the bottom of the figure, in probe translation and phrase reconstruction, the input text to be translated is parsed into sentences and each sentence is matched to a probe using a recurrent neural network (RNN) model, which uses the embedding of the sentence to predict the cluster to which it belongs. This provides a way to check the similarity of new sentences with existing cluster probes. In addition, each cluster is associated with a translation engine through a web application programming interface (API) and the sentences get translated through this translation engine. The sentences are then combined into a paragraph and provided to the user.
[0035] Note that the creation of probes may involve human experts generating a list of probes based on their experience with the language pairs under consideration.
[0036] Thus, the present invention can segment a text into multiple sentences, identify the semantic type of each sentence, and based on the best matching probe, assign the previously determined best translation engine for such corresponding probe to each sentence, while ensuring that the combination of translations from all assigned engines is a semantically and syntactically accurate translation of the entire text.
[0037] In one embodiment, as an example of a probe based on a biographical corpus, after clustering the sentence embeddings of the corpus, the following sentence is identified as the centroid of the cluster:
[0038] "Osebe has spent time learning and teaching at Clemson's university."
[0039] This sentence is used as a probe and sent to the first translator and the second translator. Figure 3 As shown, the second translator is evaluated lower than the first translator.
[0040] For a new sentence to be translated (e.g., "Victor has spent his time learning at Carnegie Mellon's University"), the invention system will select the "first translator" as the preferred translator.
[0041] Exemplary Aspects of Using a Cloud Computing Environment
[0042] Although the present detailed description includes exemplary embodiments of the invention in a cloud computing environment, it should be understood that implementations of the teachings recited herein are not limited to cloud computing environments, but can be implemented in conjunction with any other type of computing environment now known or later developed.
[0043] Cloud computing is a service delivery model for convenient, on-demand network access to a shared pool of configurable computing resources. Configurable computing resources are resources that can be quickly deployed and released with minimal management cost or interaction with the service provider, such as networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services. This cloud model can include at least five characteristics, at least three service models, and at least four deployment models.
[0044] Features are as follows:
[0045] On-demand self-service: Cloud consumers can unilaterally and automatically deploy computing capabilities such as server time and network storage on demand without human interaction with the service provider.
[0046] Broad network access: Computing power can be accessed over the network through standard mechanisms that facilitate the use of the cloud through different types of thin-client or thick-client platforms (e.g., mobile phones, laptops, personal digital assistants (PDAs)).
[0047] Resource pool: The computing resources of the provider are grouped into resource pools and serve multiple consumers through a multi-tenant model, where different physical and virtual resources are dynamically allocated and reallocated on demand. Generally, consumers cannot control or even know the exact location of the provided resources, but can specify the location (such as country, state or data center) at a higher level of abstraction, so it is location-independent.
[0048] Rapid elasticity: The ability to quickly and elastically (sometimes automatically) deploy computing power to scale up quickly, and quickly release it to scale down quickly. To consumers, the computing power available for deployment often appears to be unlimited, and any amount of computing power can be accessed at any time.
[0049] Measurable services: Cloud systems automatically control and optimize resource usage by leveraging metering capabilities at a level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency to both service providers and consumers.
[0050] The service model is as follows:
[0051] Software as a Service (SaaS): The capability provided to consumers is to use the provider's applications running on a cloud infrastructure. Applications can be accessed from various client circuits through a thin client interface such as a web browser (e.g., web-based email). Aside from limited user-specific application configuration settings, consumers neither manage nor control the underlying cloud infrastructure including networks, servers, operating systems, storage, or even individual application capabilities.
[0052] Platform as a Service (PaaS): The capability provided to consumers is to deploy applications created or acquired by consumers on the cloud infrastructure, which are created using programming languages and tools supported by the provider. Consumers neither manage nor control the underlying cloud infrastructure including networks, servers, operating systems or storage, but have control over the applications they deploy and may also have control over the configuration of the application hosting environment.
[0053] Infrastructure as a Service (IaaS): The capabilities provided to consumers are processing, storage, network, and other basic computing resources on which consumers can deploy and run arbitrary software including operating systems and applications. Consumers neither manage nor control the underlying cloud infrastructure, but have control over the operating system, storage, and applications they deploy, and may have limited control over selected network components (such as host firewalls).
[0054] The deployment model is as follows:
[0055] Private Cloud: Cloud infrastructure is run solely for an organization. The cloud infrastructure can be managed by the organization or a third party and can exist inside or outside the organization.
[0056] Community Cloud: A cloud infrastructure is shared by several organizations and supports a specific community with common interests (e.g., mission, security requirements, policies, and compliance considerations). A community cloud can be managed by multiple organizations within the community or by a third party and can exist inside or outside the community.
[0057] Public cloud: Cloud infrastructure is provided to the public or to large industry groups and is owned by an organization that sells cloud services.
[0058] Hybrid cloud: A cloud infrastructure consisting of two or more clouds (private, community or public) in a deployment model that remain distinct entities but are bound together by standardized or proprietary technologies (such as cloud burst sharing for load balancing between clouds) that enable data and application portability.
[0059] The cloud computing environment is service-oriented, with characteristics centered on statelessness, low coupling, modularity, and semantic interoperability. The core of cloud computing is the infrastructure that includes a network of interconnected nodes.
[0060] Reference now Figure 5 , which shows an example of a cloud computing node. Figure 1 The cloud computing node 10 shown is only an example of a suitable cloud computing node and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. In short, the cloud computing node 10 can be used to implement and / or perform any of the functions described above.
[0061] Although the cloud computing node 10 is described as a computer system / server 12, it can be understood that it can operate with many other general or special computing system environments or configurations. As is well known, examples of computing systems, environments and / or configurations suitable for operating with the computer system / server 12 include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop circuits, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, small computer systems, large computer systems, and distributed cloud computing technology environments including any of the above systems or circuits, etc.
[0062] Computer system / server 12 may be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. In general, program modules may include routines, programs, object programs, components, logic, data structures, etc. that perform specific tasks or implement specific abstract data types. Computer system / server 12 may be implemented in a distributed cloud computing environment where remote processing circuits linked through a communication network perform tasks. In a distributed cloud computing environment, program modules may be located on local or remote computing system storage media including storage circuits.
[0063] like Figure 1 As shown, now refer to Figure 5 The computer system / server 12 in the cloud computing node 10 is represented in the form of a general-purpose computing circuit. The components of the computer system / server 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).
[0064] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.
[0065] Computer system / server 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer system / server 12, including volatile and nonvolatile media, removable and non-removable media.
[0066] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer system / server 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 1 not shown, usually called a "hard drive"). Although Figure 1 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical medium) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one computer program product that stores a set and a plurality of program modules that include computer-readable instructions configured to perform one or more features of the present invention.
[0067] A program / utility 40 having a set (at least one) of program modules 42 may be stored in the memory 28, such program modules 42 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination thereof may be suitable for implementation in a network environment. In some embodiments, the program modules 42 are suitable for generally performing one or more functions and / or methods of the present invention.
[0068] The computer system / server 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing circuits, other peripheral devices such as a display 24, etc.) and one or more components that facilitate interaction with the computer system / server 12. Such communication may be performed via an input / output (I / O) interface 22 or any circuit that enables the computer system / server 12 to communicate with one or more other computing circuits (e.g., a network card, a modem, etc.). For example, the computer system / server 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the computer system / server 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may operate with the computer system / server 12, including but not limited to: microcode, circuit drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.
[0069] Reference now Figure 6 , in which an exemplary cloud computing environment 50 is shown. As shown, the cloud computing environment 50 includes one or more cloud computing nodes 10 with which a local computing circuit used by a cloud computing consumer can communicate, and the local computing device can be, for example, a personal digital assistant (PDA) or a mobile phone 54A, a desktop computer 54B, a laptop computer 54C and / or an automobile computer system 54N. The cloud computing nodes 10 can communicate with each other. The cloud computing nodes 10 can be physically or virtually grouped in one or more networks including but not limited to the private cloud, community cloud, public cloud or hybrid cloud or a combination thereof as described above (not shown in the figure). In this way, cloud consumers can request infrastructure as a service (IaaS), platform as a service (PaaS) and / or software as a service (SaaS) provided by the cloud computing environment 50 without having to maintain resources on the local computing circuit. It should be understood that Figure 6 The various types of computing circuits 54A-N shown are merely illustrative, and the cloud computing node 10 and the cloud computing environment 50 can communicate with any type of computing circuit on any type of network and / or network addressable connection (eg, using a web browser).
[0070] Reference now Figure 7 , which shows a cloud computing environment 50 ( Figure 6 ) provides an exemplary set of functional abstraction layers. First, it should be understood that Figure 7 The components, layers and functions shown are only exemplary, and the embodiments of the present invention are not limited thereto. As shown in the figure, the following layers and corresponding functions are provided:
[0071] The hardware and software layer 60 includes hardware and software components. Examples of hardware components include: host 61; server 62 based on RISC (Reduced Instruction Set Computer) architecture; server 63; blade server 64; storage circuit 65; network and network components 66. Examples of software components include: network application server software 67 and database software 68.
[0072] The virtualization layer 70 provides an abstraction layer that can provide examples of the following virtual entities: virtual servers 71 , virtual storage 72 , virtual networks 73 (including virtual private networks), virtual applications and operating systems 74 , and virtual clients 75 .
[0073] In one example, the management layer 80 may provide the following functions: Resource provisioning function 81: providing dynamic acquisition of computing resources and other resources for performing tasks in a cloud computing environment; Metering and pricing function 82: tracking the cost of resource usage within the cloud computing environment and providing bills and invoices for this purpose. In one example, the resource may include an application software license. Security function: providing identity authentication for cloud consumers and tasks, and providing protection for data and other resources. User portal function 83: providing access to the cloud computing environment for consumers and system administrators. Service level management function 84: providing allocation and management of cloud computing resources to meet the required service levels. Service level agreement (SLA) planning and fulfillment function 85: providing pre-arrangement and provision for future demand for cloud computing resources predicted according to the SLA.
[0074] The workload layer 90 provides examples of functions that may be implemented in a cloud computing environment. In this layer, examples of workloads or functions that may be provided include: mapping and navigation 91; software development and lifecycle management 92; virtual classroom teaching provision 93; data analysis processing 94; transaction processing 95; and the translation engine recommendation method 100 according to the present invention.
[0075] At any possible technical detail combination level, the present invention may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.
[0076] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.
[0077] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.
[0078] The computer program instructions for performing the operation of the present invention may be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present invention.
[0079] Various aspects of the present invention are described herein with reference to the flow charts and / or block diagrams of the methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each box of the flow chart and / or block diagram and the combination of each box in the flow chart and / or block diagram can be implemented by computer-readable program instructions.
[0080] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0081] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0082] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.
[0083] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
[0084] Furthermore, it is the applicant's intent to encompass equivalents of all claim elements, and any amendment of any claim in this application should not be construed as a waiver of any benefit or right to equivalents of any element or feature of the amended claim.
Claims
1. A computer-implemented translation engine suggestion method, the method comprising: Split parts of the input text into segments; dividing the input text into segments according to the segmentation; identifying a plurality of combinations, wherein each combination includes a probe of a third-party translation engine that has a highest semantic similarity to each segment of the input text and that is an accurate translation of the segment; applying each segment to each of the combinations of the probe and the third-party translation engine to verify a best combination among the plurality of combinations for the segment that results in a best translation of the segment from the probe of the third-party translation engine compared to other combinations applied to the same segment; and A translation of the input text is output by combining the best translation from each of the combination of the probes and the third party translation engine that results in the best combination for each segment.
2. The method of claim 1, wherein: The process of learning segments for segmenting the input text and training the model to recognize probes is performed by processing the input training text by tokenizing the text into sentences mapped to a vector space using sentence embeddings.
3. The method of claim 2, wherein the sentences are clustered based on the sentence embeddings, and in, The sentences at the center of the cluster are identified and assigned a cluster label.
4. The method of claim 3 further comprising querying a human expert to translate the sentence at the centroid into a target language.
5. The method of claim 4, wherein the step of applying the identified probes to each segment using a corresponding third-party translation engine is compared with a human expert translation.
6. The method of claim 5, wherein the third-party translation engine performs a score based on the comparison result, and in, The application is the third-party translation engine with the highest rating for each segment application.
7. The method of claim 1, embodied in a cloud computing environment.
8. A computer program product for metadata-based characterization, the computer program product comprising a computer-readable storage medium having program instructions contained therein, the program instructions being executable by a computer to cause the computer to perform the method of any one of claims 1 to 7.
9. A metadata-based characterization system, the system comprising: processor; as well as A memory storing instructions for causing a processor to execute the method according to any one of claims 1 to 7.
10. A metadata-based characterization system, comprising a module for executing each step of the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Evaluating query translations for cross-language query suggestion
CN104011712A
Input method and device oriented at computer-assisting translation
CN104462072A
Machine translation apparatus and machine translation method
CN105468585A
A cyclic transfer learning method for establishing a word translation model
CN109388809A
Distributed selection and fusion of translation tools
CN109558602A
