2Know Magazine: Sharing KM Knowledge
2Know: Sharing KM Knowledge
January 2022 - Magazine No. 268
January 2022 - Magazine No. 268

Natural Language Processing (NLP) is a sub-genre of AI and linguistics which deals with issues regarding the comprehension, interpretation and processing of language to make computers “understand” content said or written in human language.

An outline of the problem

When adapting the language and machine learning to Hebrew, it was found that converting speech in Hebrew to text is performed reasonably; the complex problem they face is the difficulty in deciphering the meaning of texts:

Due to the Hebrew syntax and grammar differ greatly from those of languages currently used in NLP modules, one cannot use existing modules cannot be used directly and expect satisfying results. For example, Hebrew does not have any vowels and three or four-lettered words can be read in several different ways. Furthermore, unlike English, the order of the words in a sentence in Hebrew is irrelevant to its meaning.

The machines also struggle to understand the meaning of sentences and place words in their correct context.

Since there are relatively very few Hebrew speakers, despite some of these companies’ development centers being located in Israel, the economic justification for investing in this issue is clearly insufficient.  

Attempts at a solution

One way to solve the problem was implemented by the ICT Authority, which began in 2020 to actively create a database to be used to train machines to understand the Hebrew language, that will be available to governmental services, start-ups, and large corporations. The database – “a tagged manual corpus of Contemporary Hebrew” ­­– is a database of sentences in Hebrew defined by its Dictionary entry, parts of speech (nouns, verbs, etc.), and syntactic entities (subject of sentence, etc.).

After feeding many sentences into it via machine learning, the program will be able to begin to address sentences that weren’t previously fed into it. In 2021, the project won a special certificate of appreciation by the panel of judges at the Bureau of Information Technology in Israel conference.

Another approach researchers implement, in order to develop Hebrew-speaking voice assistants (like the English-speaking Alexa and Siri) is developing computerized modules based on Big Data. These modules process words in context, in relation to other words preceding and following them in the sentence, and are based on processing large amounts of data. One example of this is Google’s Bert (Bidirectional Encoder Representation from Transformers) launched in 2019. BERT helps Google with NLP and by simulating an artificial neuron network it allows a better understanding of the query’s different words’ context, especially in cases of double meaning. One of the innovations BERT offers is the ability to understand the links between the words near – but not necessarily immediately adjacent – words. In other words, it can even guess missing words.

In conclusion, despite objective difficulties derived from the properties of Hebrew/Semitic languages compared to other languages, it seems that with technology’s accelerated advance nowadays, and hopefully considering that besides approximately ten million Hebrew speakers there about three hundred million Arabic speakers, we will be able to soon enjoy voice/textual chatbots that fully and sufficiently support in Hebrew, as well.

The field of search engines is evolving in recent years, especially due to the emergence of AI applications, which signify a huge leap in search engines nowadays and in the future.

Google, for example, uses a “Knowledge Graph”, which is the target base based on entities rather than key words which enables presenting a “information card” as an answer to search queries which features a bast amount of data semantically related to the entity which the search was conducted on. Google builds the “Knowledge Graph” based on analyzing previous queries and links between web pages, analyzing databases such as Wikipedia, Wikimedia, and the CIA’s The World Factbook, Schema.org and other reliable websites.


Its objective is to create quick answers to users so that they may use this data to solve their query without searching the internet for an answer, thus reducing the search time by presenting the answer in the search page.

For example, searching for the term AI on Google displays the term’s meaning before presenting the search results using the Knowledge Graph.

In these cases, the AI assists the search engine in understanding the person’s natural language by recognizing the meaning related to the query and the information required to provide answers. Another improvement is the development of search engines’ ability to understand the user’s intent behind the search query. The challenging aspect of this issue is derived from the fact that we use search engines for various purposes (shopping, research, data detection, etc.) and since there are many cases in which one key word can bear several meanings, by analyzing clicking patterns and types of content that the user has searched previously, the search engine can leverage machine learning to determine the searcher’s intent.

For example, Google uses BERT (Bidirectional Encoder Representations from Transformers), which is an NLP tool Google uses to better understand the context of the user’s search query and the nuance and connection between the words. This is done by deciphering and studying the users’ interaction with the content. Furthermore, Google uses RankBrain, a machine learning algorithm it developed, which not only assists in detecting patterns in user queries, but also assists the search engine in predicting which search result the user will click on when presented a query not previously entered based on analyzing historical search results. RankBrain is also used for finding synonyms so that user search results can include only the relevant synonyms and not the one entered by the user.



Other search elements rapidly expanded are voice searches and visual searches:

A voice search: with the emergence of “virtual assistants” such as Siri and Alex, voice queries have become increasingly populat among search engines. In 2020, Google reported that 20% of searches via smartphones are voice queries. These have become longer and more complex. In these cases, NLP can assist us in validating the content’s quality, and the search engines’ abilities to [process content of its kind.

Visual search: every second a sum total of 5000 pictures on social networks such Facebook and Instagram. Hundreds of millions of pictures are uploaded to them on a daily basis. AI machine learning tools enable an analysis of color and shape patterns and attach them to all data on the picture, in order to assist the search engine in understanding the picture and its meaning. For example, the Google search engine is able to not only catalog pictures for picture searches, but also allows users to find other appearances of this picture on the internet, as well as similar pictures of similar themes or color palates and data on the themes displayed in it. Furthermore. The picture search can be reversed, enabling users to search using a picture rather than a text query, using apps such as Google Lens, which will become in the near future a dominant marketing trend in the field of search engines.

Another use of AI that can improve search engine results is the use of machine learning to identify templates that assist in detecting spam or copied content. Machine learning identifies these patterns by learning from similar past incidents to recognize the use of new templates and spam techniques, then label them as problematic.

For example, despite Google still using people to QA its content, using machine learning tools to detect these patterns drastically reduces the amount of manpower this task now requires.

In conclusion, AI technologies such as NLP machine learning are slowly changing the way search engines are finding and rating data. The more users use them, the “smarter” and more accurate will the received results be. However, it is worth noting that machine learning will never be perfect, and that there will always be a need for human workers to perform the fine-tuning and improvements companies make to their search engines.











Written By Michal Blumenfeld Sagi

Most of our lives can be divided into ‘work hours,’ during which we are serious and professional, and ‘free time.’ We are usually less formal and looser during these hours. However, what if our organization offered us some magic and enjoyment as part of work time?


Organizations have begun to recognize that gamification is not child’s play in recent years. It benefits the organization. Among the advantages of implementing gamification in organizational processes are: learning with enjoyment, increasing knowledge sharing in the organization, enhancing camaraderie and collaboration among workers, enhancing values, belonging and connection to the organization, empowering workers by making them the heroes of the game, encouraging ‘out of the box thinking, improving leading and problem-solving skills, and designing a success experience by introducing competitive elements while retaining a sense of success among all participants.


How do you incorporate gamification in an organization? There are infinite ways to gamify most processes in the organization. Gamification can be physical, digital, or hybrid. All we need to do is think in and outside of the game’s box.


Here are a few ways to integrate gamification in Knowledge Management processes:


  • Who likes debriefing? While this process is vital to both the organization and its workers, most of us could quickly produce more enjoyable task​, ​gamifying the processcan enhance positive emotions towards it and is an excellent platform for role-playing. For example, we can create an interactive scene of the reviewed case. Thus, we can play detectives, using clues to discover what happened. We can play “Clue” using a fixed database of relevant questions.
  • Let us retain knowledge using an insight database: this activity can complement lessons learned processes or stand along. For example, we can design a game that involves finding clues in specific insights, have workers compete in generating correct insights from case descriptions (independent formulation or choosing from a database containing correct and incorrect insights), and many more.
  • Changes are challenging, but who does not enjoy a challenging game? Incorporating gamification into change processes can make them more manageable, more pleasant, and more enjoyable experiences. We can design a gamified activity (for example, “Who Moved My Cheese?”) to infuse the process with humor and lightness. This will make the difficulties of change an exciting adventure.
  • Special events: it is highly advisable to create gamified activities revolving around special events in the organization, such as team building days or professional conventions, and incorporate content and value with the enjoyment. You can create game templates that suit workshops or professional conventions on themes such as creativity, innovation, and thinking outside the box.


Those are just some suggestions. This is where we need to express our creativity while remembering that gamification must suit the organization’s needs and resources. Our activities can be complex and intricate, but simple actions may be more effective if we remain focused on the value we wish to generate. To do this, we must not forget the three basic rules of gamification: fun, fun, and fun.



Written by Rom Knowledgeware
Fax 077-5020772 * Tel 077-5020771/3 * Bar Kochva 23 st., Bnei Brak Postal: 67135